Battery state estimating system and battery state estimating method

Abstract

A battery state estimating method is proposed, and includes performing a battery measuring step, a heat generation estimating step, a surface temperature observing step, a sliding mode step, an adaptive controlling step and a battery state calculating step. The surface temperature observing step includes observing a battery heat generation, a surface temperature measurement value and an ambient temperature measurement value to establish a surface temperature observation model. The sliding mode step includes defining a sliding function, a plurality of parameter error equations and a control input equation based on a surface temperature error value. The adaptive controlling step includes respectively updating a plurality of adaptive parameters and an adaptive control item in the surface temperature observation model through the parameter error equations and the control input equation, and making the parameter error equations to converge within a finite time to generate a plurality of estimated parameters. The battery state calculating step includes calculating the estimated parameters to estimate the battery state. Therefore, the core temperature value and thermal resistance of the battery can be estimated in real time to avoid thermal runaway of the battery.

Description

[Technical Field]

[0001] This elucidation is about a battery state estimation system and a battery state estimation method, in particular, a battery state estimation system and battery state estimation method for estimating core temperature values, core-to-surface thermal resistance and surface-to-ambient thermal resistance. [Previous Technology]

[0002] In recent years, the gradual reduction of petrochemical fuels has gradually attracted attention to the issues of global warming and energy shortages. The sales of electric vehicles are increasing due to their advantages of low pollution and high energy use efficiency. However, some niche electric vehicles, such as agricultural electric vehicles, pose a high threat to battery safety due to their rather harsh usage scenarios. In addition, the growing demand for efficient and stable energy storage solutions has led to significant advances in Lithium-ion Battery (LIB). Despite their high energy density, long cycle life, and low self-discharge rate, lithium-ion batteries face huge thermal management challenges, such as overheating and thermal runaway (Thermal Runaway). If the battery status cannot be accurately estimated, it can lead to catastrophic vehicle failure and even cause safety issues for driving and passengers in the vehicle. It is crucial to ensure the performance, safety and longevity of thermal management of batteries.

[0003] However, the Battery Management System (BMS) can only monitor the surface temperature of the battery, while the core temperature value cannot be obtained through non-destructive measurements. Furthermore, existing monitoring software or algorithms cannot accurately and dynamically monitor battery-related thermal parameters, such as thermal resistance from the environment to the battery surface and thermal resistance from the core to the battery surface. It follows that there is currently a lack of battery status estimation system and battery status estimation method that can be immediately, accurately and can be measured with non-destructive measures in the market. [Invention Contents]

[0004] Therefore, the purpose of this elucidation is to provide a battery state estimation system and a battery state estimation method that adopts an Adaptive Control algorithm to model the thermal behavior of the battery by estimating the core temperature value, core-to-surface thermal resistance and surface-to-environment thermal resistance of the battery. The system and method of this elucidation of the content can not only know how the battery heat is dissipated from the battery core to the battery surface, but also maintain the thermal balance to ensure the battery operation in the optimal temperature range, thereby improving the battery's performance, safety and service life, and supporting applications in portable electronics, electric vehicles and renewable energy storage.

[0005] According to one embodiment of the structural pattern of this disclosure, a battery state estimation system is provided, which includes a measuring device and a processing device. The measuring device is used to measure a battery charging / discharging current, a battery terminal voltage, a surface temperature measurement value, and a peripheral temperature measurement value of a battery. The processing device is connected to the measuring device and obtains the battery charging / discharging current, battery terminal voltage, surface temperature measurement value, and peripheral temperature measurement value. The processing device is configured to perform the following operations: performing a heat generation estimation operation on the battery charging / discharging current and battery terminal voltage to generate battery heat; performing a surface temperature observation operation, wherein the surface temperature observation operation includes observing battery heat, surface temperature measurements, and ambient temperature measurements to establish a surface temperature observation model and generate a surface temperature error value, the surface temperature observation model including complex adaptive parameters and an adaptive control term; performing a sliding mode operation on the surface temperature error value to define a sliding function, and defining complex parameter error equations and a control input equation based on the sliding function; performing an adaptive control operation, wherein the adaptive control operation updates the adaptive parameters and adaptive control term through these parameter error equations and the control input equation, respectively, and stabilizes the sliding function within a finite time, so that these parameter error equations converge within a finite time to generate complex estimated parameters; and performing a battery state calculation operation on these estimated parameters to estimate a battery state.

[0006] Other embodiments of the aforementioned implementation are as follows: The aforementioned heat generation estimation operation includes an internal resistance estimation operation and a heat generation calculation operation. The internal resistance estimation operation includes estimating a battery internal resistance value based on the battery charging / discharging current and the battery terminal voltage. The heat generation calculation operation includes calculating the battery heat generation by combining the battery internal resistance value and the battery charging / discharging current.

[0007] Other embodiments of the aforementioned implementation are as follows: The aforementioned surface temperature observation model further includes battery heat generation and a surface temperature estimate, and conforms to the following formula:

[0008] where is the estimated surface temperature value, is the first derivative of the estimated surface temperature value, is the second derivative of the estimated surface temperature value, , , and are these adaptive parameters respectively, is the adaptive control term, and is the battery heat generation. The surface temperature observation operation further includes calculating the surface temperature error value by comparing the estimated surface temperature value with the measured surface temperature value.

[0009] Other embodiments of the foregoing implementation are as follows: The foregoing sliding function includes a surface temperature error value and a control gain, and conforms to the following formula:

[0010] Where is the sliding function, and is the surface temperature error value. The gain is controlled by differentiating the surface temperature error value. The sliding mode operation involves driving these parameter error equations into a sliding mode. When these parameter error equations enter the sliding mode, the sliding function reaches a sliding surface, which is 0. The sliding mode operation further includes confirming that these parameter error equations have entered the sliding mode through a Lyapunov function.

[0011] Other embodiments of the foregoing implementation are as follows: The aforementioned battery state calculation operation includes a battery thermal resistance calculation operation and a core temperature calculation operation. The battery thermal resistance calculation operation includes calculating these estimated parameters to generate a surface-to-ambient thermal resistance or a core-to-surface thermal resistance for one battery state. The core temperature calculation operation includes calculating these estimated parameters to generate a core temperature value for one battery state.

[0012] According to one embodiment of the method of the present disclosure, a battery state estimation method is provided, which includes: measuring a battery charging and discharging current, a battery terminal voltage, a surface temperature measurement value, and a peripheral temperature measurement value of a battery using a measuring device; performing a heat generation estimation operation on the battery charging and discharging current and the battery terminal voltage using a processing device to generate battery heat generation; and performing a surface temperature observation operation using the processing device, wherein the surface temperature observation operation includes observing the battery heat generation, the surface temperature measurement value, and the peripheral temperature measurement value to establish a surface temperature observation model and generate a surface temperature error value, the surface temperature observation model including complex adaptation. The system comprises: a parameter and an adaptive control term; a sliding function defined by performing a sliding mode operation on the surface temperature error value using a processing device; a complex parameter error equation and a control input equation defined based on the sliding function; an adaptive control operation performed by the processing device, wherein the adaptive control operation updates the adaptive parameters and adaptive control term respectively through these parameter error equations and the control input equation, and stabilizes the sliding function within a finite time, so that these parameter error equations converge within a finite time to generate complex estimated parameters; and a battery state calculation operation performed by the processing device on these estimated parameters to estimate a battery state.

[0013] Other embodiments of the aforementioned implementation are as follows: The aforementioned heat generation estimation operation includes an internal resistance estimation operation and a heat generation calculation operation. The internal resistance estimation operation includes estimating a battery internal resistance value based on the battery charging / discharging current and the battery terminal voltage. The heat generation calculation operation includes calculating the battery heat generation by combining the battery internal resistance value and the battery charging / discharging current.

[0014] Other embodiments of the foregoing implementation are as follows: The foregoing surface temperature observation model further includes battery heat generation and a surface temperature estimate, and conforms to the following formula:

[0015] where is the estimated surface temperature value, is the first derivative of the estimated surface temperature value, is the second derivative of the estimated surface temperature value, , , and are these adaptive parameters respectively, is the adaptive control term, and is the battery heat generation. The surface temperature observation operation further includes calculating the surface temperature error value by combining the estimated surface temperature value and the measured surface temperature value.

[0016] Other embodiments of the foregoing implementation are as follows: The foregoing sliding function includes a surface temperature error value and a control gain, and conforms to the following formula:

[0017] Where is the sliding function, and is the surface temperature error value. The gain is controlled by differentiating the surface temperature error value. The sliding mode operation involves driving these parameter error equations into a sliding mode. When these parameter error equations enter the sliding mode, the sliding function reaches a sliding surface, which is 0. The sliding mode operation further includes confirming that these parameter error equations have entered the sliding mode through a Lyapunov function.

[0018] Other embodiments of the foregoing implementation are as follows: The aforementioned battery state calculation operation includes a battery thermal resistance calculation operation and a core temperature calculation operation. The battery thermal resistance calculation operation includes calculating these estimated parameters to generate a surface-to-ambient thermal resistance or a core-to-surface thermal resistance for one battery state. The core temperature calculation operation includes calculating these estimated parameters to generate a core temperature value for one battery state.

Implementation Method

[0019] Several embodiments of this disclosure will now be described with reference to the accompanying drawings. For clarity, many practical details will be described in the following description. However, it should be understood that these practical details should not be used to limit the disclosure. That is, in some embodiments of this disclosure, these practical details are not essential. Furthermore, for the sake of simplicity, some conventional structures and elements will be shown in the drawings in a simple schematic manner; and repeated elements may be represented by the same number.

[0020] Furthermore, in this document, when a component (or unit or module, etc.) is "connected" to another component, it can mean that the component is directly connected to the other component, or it can mean that the component is indirectly connected to the other component, that is, there is another component between the component and the other component. Only when it is explicitly stated that a component is "directly connected" to another component does it mean that there is no other component between the component and the other component. The terms "first," "second," "third," etc., are only used to describe different components and do not limit the components themselves. Therefore, the first component can also be referred to as the second component. Moreover, the combinations of components / units / circuits in this document are not combinations that are generally known, conventional, or customary in this field. Whether the components / units / circuits themselves are customary cannot be used to determine whether their combination relationship is easily completed by someone of ordinary knowledge in the art.

[0021] This disclosure proposes an adaptive control algorithm for managing battery thermal behavior, and implements and verifies it through Simulink simulation. The field of adaptive control theory provides a potential framework for creating controllers. Controllers can dynamically modify their parameters to respond to the behavior of the controlled system, and employ adaptive control algorithms to develop more responsive and efficient thermal management strategies, dynamically adapting to different operating conditions, thereby ensuring the battery operates within its optimal temperature range, thus improving performance and safety. Specifically, the adaptive control algorithm of this disclosure can combine key thermal parameters in the battery state, such as core temperature, core-to-surface thermal resistance, and surface-to-ambient thermal resistance, to estimate and maintain optimal thermal conditions, representing how battery heat dissipates from the battery core to the battery surface, maintaining thermal equilibrium and preventing dangerous temperature peaks. The goal is to model the battery thermal behavior and verify the control system.

[0022] Please refer to Figures 1 and 2, where Figure 1 is a block diagram illustrating the battery state estimation system 100 of the first embodiment of this disclosure, and Figure 2 is a schematic diagram illustrating the thermal parameters related to the battery 110 in the battery state estimation system 100 of Figure 1. As shown in Figures 1 and 2, the battery state estimation system 100 includes a battery 110, a measuring device 200, and a processing device 300. The measuring device 200 is used to measure a battery charge / discharge current, a battery terminal voltage, a surface temperature measurement, and a peripheral temperature measurement of the battery 110. The processing device 300 is electrically connected to the measuring device 200 and obtains the battery charge / discharge current, battery terminal voltage, surface temperature measurement, and peripheral temperature measurement. The processing device 300 can be configured to perform operations including: a heat generation estimation operation 310, a surface temperature observation operation 320, a sliding mode operation 330, an adaptive control operation 340, and a battery state calculation operation 350.

[0023] The processing device 300 performs a heat generation estimation operation 310 on the battery charging and discharging current and battery terminal voltage to generate battery heat. The surface temperature observation operation 320 includes observing the battery heat, surface temperature measurement value, and ambient temperature measurement value to establish a surface temperature observation model and generate a surface temperature error value, wherein the surface temperature observation model includes complex adaptive parameters and an adaptive control term. The processing device 300 performs a sliding mode operation 330 on the surface temperature error value to define a sliding function, and defines a complex parameter error equation and a control input equation based on the sliding function. The adaptive control operation 340 updates the adaptive parameters and adaptive control term in the surface temperature observation model through these parameter error equations and control input equations, and stabilizes the sliding function within a finite time to make these parameter error equations converge within a finite time to generate complex estimated parameters. Then, the processing device 300 performs a battery state calculation operation 350 on these estimated parameters to estimate the battery state of battery 110.

[0024] Specifically, the battery 110 may be, but is not limited to, a lithium-ion battery (LIB), and the measuring device 200 may be, for example, an electronic device used to measure properties of the battery 110 such as current, voltage, or temperature. In Figure 2, the measuring device 200 can obtain surface temperature measurement values ​​and ambient temperature measurement values ​​by measuring the surface temperature of the battery 110 and the temperature of the surrounding area of ​​the battery 110, respectively. In addition, the battery state of the battery 110 may include a core temperature value, a core-to-surface thermal resistance, and a surface-to-ambient thermal resistance, wherein the core temperature value cannot be directly obtained by non-destructive measurement.

[0025] The processing device 300 may include a memory and a processor. The memory may include random access memory (RAM) or other types of dynamic storage devices capable of storing information, program code, and instructions for execution by the processor. Software programs or software algorithms corresponding to the heat generation estimation operation 310, surface temperature observation operation 320, sliding mode operation 330, adaptive control operation 340, and battery state calculation operation 350 are stored in the memory. The processor may be a cloud processor, digital signal processor (DSP), microprocessor (MPU), central processing unit (CPU), or other electronic processor, but this disclosure is not limited to the above. The processor is electrically connected to the memory and configured to access the information, program code, and instructions in the memory to automatically implement the battery state estimation method proposed in this disclosure.

[0026] The battery state estimation system 100 disclosed herein employs an adaptive control algorithm to estimate core temperature, core-to-surface thermal resistance, and surface-to-ambient thermal resistance—all crucial to thermal management—based on measurements of battery charging / discharging current, battery terminal voltage, surface temperature, and ambient temperature obtained from the measuring device 200. During dynamic operation of the battery 110, these three thermal parameters are variables that change with temperature and battery state. This disclosure establishes a physical estimation system based on the adaptive control algorithm to estimate these dynamic thermal parameters. The most significant feature of using the adaptive control algorithm for thermal parameter estimation is that the surface temperature observation operation 320, the sliding mode operation 330, and the adaptive control operation 340 can be considered as an adaptive observer. Utilizing existing instantaneous operating current and voltage measurements under dynamic conditions and temperature measurements detected non-destructively, it can automatically track and correct to the correct thermal parameters. The following paragraphs, with accompanying diagrams, will explain in detail the operation of each step in the battery state estimation method disclosed herein.

[0027] Please refer to Figures 1, 2, 3, and 4, wherein Figure 3 is a flowchart illustrating the battery state estimation method S0 of the second embodiment of the present disclosure, and Figure 4 is a flowchart illustrating the various operations in the battery state estimation method S0 of Figure 3. As shown in Figures 1, 2, 3, and 4, the battery state estimation method S0 can be automatically executed by the battery state estimation system 100, and includes the following steps S01, S02, S03, S04, S05, and S06, and steps S01, S02, S03, S04, S05, and S06 are executed sequentially.

[0028] Step S01 includes measuring the battery charging / discharging current, battery terminal voltage, surface temperature, and ambient temperature of the battery 110 using the measuring device 200. Step S02 includes generating battery heat by performing a heat generation estimation operation 310 on the battery charging / discharging current and battery terminal voltage using the processing device 300. Step S03 includes performing a surface temperature observation operation 320 using the processing device 300, wherein the surface temperature observation operation 320 includes observing the battery heat generation, surface temperature measurement, and ambient temperature measurement to establish a surface temperature observation model and generate a surface temperature error value. Step S04 includes defining a sliding function by performing a sliding mode operation 330 on the surface temperature error value using the processing device 300, and defining a complex parameter error equation and a control input equation based on the sliding function. Step S05 includes executing an adaptive control operation 340 by the processing device 300. The adaptive control operation 340 updates the complex adaptive parameters and adaptive control terms in the surface temperature observation model through these parameter error equations and control input equations, respectively, and stabilizes the sliding function within a finite time, so that these parameter error equations converge within a finite time to generate complex estimated parameters. Step S06 includes performing a battery state calculation operation 350 on these estimated parameters by the processing device 300 to estimate the battery state of the battery 110. Therefore, the battery state estimation method S0 of this disclosure uses an adaptive control algorithm to model the thermal behavior of the battery 110, thereby estimating the core temperature value, core-to-surface thermal resistance, and surface-to-ambient thermal resistance of the battery 110. Therefore, the battery state estimation system 100 and battery state estimation method S0 disclosed herein can not only determine how the battery heat dissipates from the battery core to the battery surface, but also maintain thermal balance to ensure that the battery 110 operates within the optimal temperature range, thereby improving the performance, safety and lifespan of the battery 110, and supporting applications in portable electronic products, electric vehicles and renewable energy storage.

[0029] The following examples, based on the second embodiment, illustrate two practical cases to demonstrate real-world applications. It should be noted that the battery state estimation method S0 disclosed herein allows for setting initial conditions for control gain, adaptive gain, and adaptive parameters in both cases of tracking surface-to-ambient thermal resistance and core-to-surface thermal resistance. This allows for fine-tuning of the adaptive algorithm and ensures that the system starts from a clearly defined state, facilitating accurate tracking and adjustment during simulation. Control and adaptive gains are crucial for adjusting adaptive control, ensuring the system responds appropriately to changing thermal conditions.

[0030] 〈Case 1〉: Tracing the surface-to-environment thermal resistance.

[0031] In step S01, the measuring device 200 detects the battery 110 and inputs the battery charging and discharging current, battery terminal voltage, surface temperature measurement value and ambient temperature measurement value to the processing device 300.

[0032] In step S02, the heat generation estimation operation 310 may include an internal resistance estimation operation 311 and a heat generation calculation operation 312 executed sequentially. The internal resistance estimation operation 311 includes estimating a battery internal resistance value based on the battery charging / discharging current and the battery terminal voltage. The heat generation calculation operation 312 includes calculating the battery heat generation by combining the battery internal resistance value and the battery charging / discharging current. The method of estimating the battery internal resistance value based on the battery charging / discharging current and the battery terminal voltage is prior art and is not the focus of this disclosure; details will not be elaborated further. In other embodiments, the battery internal resistance value may also be a known battery parameter and stored in the memory of the processing device.

[0033] Before performing the surface temperature observation operation 320 in step S03 to establish the surface temperature observation model, by analyzing the energy balance equations of the battery core and surface, this disclosure can first derive the relationship that helps to estimate the core-to-surface thermal resistance and implement effective thermal management strategies. The core battery energy equation can conform to the following equation (1): (1).

[0034] In formula (1), is the core thermal capacity of battery 110, is the measured surface temperature, is the core temperature value, is the rate of change of the core temperature value (i.e., the first derivative of the core temperature value), is the thermal resistance from the core to the surface, and is the heat generated by the internal resistance of the battery and is defined as the following formula (2): (2).

[0035] In equation (2), is the internal resistance of the battery, and is the charging and discharging current of the battery. On the other hand, the energy equation of the surface battery can conform to the following equation (3): (3).

[0036] In equation (3), is the surface heat capacity of battery 110, is the rate of change of the surface temperature measurement (i.e., the first derivative of the surface temperature measurement), is the ambient temperature measurement, and is the surface-to-environment thermal resistance. By combining the core and surface battery energy equations, this disclosure can derive that the relationship between the core temperature value, the surface temperature measurement value, and the ambient temperature measurement value conforms to the following equation (4): (4).

[0037] After differentiating equation (4) and substituting it into equation (1) of the aforementioned core battery energy equation, the following equation (5) can be obtained. In order to omit the core temperature value that cannot be detected by non-destructive measurement, equation (5) is simplified to derive the following equation (6): (5); (6).

[0038] In formula (6), , and are all model parameters, and can respectively conform to the following formulas (7), (8), and (9): (7); (8); (9).

[0039] It is worth mentioning that these three model parameters need to be estimated so that the surface-to-environment thermal resistance can be obtained from the estimated and derived parameters, as shown in the following formula (10): (10).

[0040] In this way, the surface temperature observation operation 320 can model equation (6) based on the sliding mode observer. In step S03, the surface temperature observation operation 320 establishes a surface temperature observation model by observing the battery heat generation, surface temperature measurement value and surrounding temperature measurement value. In particular, the surface temperature observation model is a quadratic differential estimation model, which includes complex fitness parameters, adaptive control terms, surface temperature estimation value and battery heat generation, and conforms to the following equation (11): (11).

[0041] In equation (11), is the estimated surface temperature value, is the first derivative of the estimated surface temperature value, is the second derivative of the estimated surface temperature value, and are these adaptive parameters, is the adaptive control term, and is the battery heat generation. In detail, (i.e., , and ) can be regarded as the estimated values ​​corresponding to the aforementioned model parameters, that is, the estimated values ​​of and in equation (6). The adaptive control term is mainly used to correct the model error between the surface temperature observation model and equation (6) by utilizing the error between the estimated value and the actual measurement value. In other words, equation (6) can be a correct model of the actual measurement, the surface temperature observation model of equation (11) is an observer of equation (6), and in equation (11) is the estimated value in equation (6).

[0042] In addition, the surface temperature observation operation 320 may further include calculating the surface temperature estimate and the surface temperature measurement to generate a surface temperature error value. Based on the model error, the surface temperature observation operation 320 may define and, therefore the processing device 300 may rewrite equation (11) as the following equation (12): (12).

[0043] In order to reduce the order of the second derivative of the surface temperature error value in equation (12), the sliding mode operation 330 defines a sliding function based on the surface temperature error value, and then defines a complex parameter error equation and a control input equation based on the sliding function. The sliding function includes the surface temperature error value and a control gain, and can satisfy the following equation (13). In addition, the sliding mode operation 330 can include taking the first derivative with respect to the sliding function to generate the following equation (14): (13); (14).

[0044] In equations (13) and (14), and are sliding functions and their first derivatives, is the surface temperature error value, the surface temperature error value is differentiated once, the surface temperature error value is differentiated twice, and the gain is controlled.

[0045] In step S04, the sliding mode operation 330 may further include driving these parameter error equations into sliding mode. When these parameter error equations enter sliding mode, the sliding function reaches a sliding surface, which is 0 (i.e.). Furthermore, the sliding mode operation 330 may further include confirming that these parameter error equations have entered sliding mode through a Lyapunov function. In other words, in order to satisfy the Lyapunov stability verification, it is necessary to ensure that the sliding function satisfies the Lyapunov function. The Lyapunov function and its first derivative may satisfy the following equations (15) and (16): (15); (16).

[0046] In equations (15) and (16), and are the Lyapunov functions and their first derivatives, is the first derivative of the adaptive parameter (i.e. ), is the parameter error value between the adaptive parameter and the model parameter, and is the adaptive gain.

[0047] Based on the dynamics of the alternative sliding surface, equation (16) is rewritten as the following equation (17): (17).

[0048] The control input equation used to update the adaptive control term can be used as a control law. In order to satisfy the Lyapunov stability verification, the control input equation must satisfy the following condition (18): (18).

[0049] In equation (18), both are control gains. Furthermore, in order to satisfy that the first derivative of the Lyapunov function is less than zero (equivalent to being negative), thus driving convergence, the parameter error equation used to update the adaptive parameters is derived, and it conforms to the following equations (19), (20), and (21): (19); (20); (21).

[0050] In equations (19), (20), and (21), (i.e., and) represent the adaptive gain. In step S05, the adaptive control operation 340, based on the adaptive rule, ensures that the adaptive parameters and adaptive control terms are updated with iterative calculations, causing the adaptive parameters to converge to the model parameters of the true model. In other words, equations (19), (20), and (21) will be continuously updated and converge to equations (7), (8), and (9). The sliding function defined by the sliding mode operation 330 tends to stabilize, and the processing device 300 can derive the estimated parameters corresponding to the convergence of the adaptive parameters to the model parameters through the parameter error equation.

[0051] In step S06, the battery state calculation operation 350 may include a battery thermal resistance calculation operation 351. The battery thermal resistance calculation operation 351 includes calculating the surface-to-ambient thermal resistance of the battery state by performing complex estimated parameters (i.e., AND) calculations. That is, the surface-to-ambient thermal resistance can be derived from the estimated AND, as shown in equation (10). It can be seen that the control strategy of the adaptive control algorithm ensures the stability and accuracy of the surface temperature observation model, thereby enabling precise estimation and control of the thermal behavior of the battery 110.

[0052] 〈Case 2〉: Tracking core temperature value and core-to-surface thermal resistance.

[0053] Step S01 is the same as step S02 and Case 1. Before performing the surface temperature observation operation 320 in step S03 to establish the surface temperature observation model, this disclosure can also first derive the relationship that helps to estimate the core-to-surface thermal resistance and implement effective thermal management strategies by analyzing the energy balance equations of the battery core and surface. The core battery energy equation is the same as equation (1), while the surface battery energy equation is modified to conform to the following equation (22): (22).

[0054] By combining the energy equations of the core and surface cells, this disclosure can derive the relationship between the core temperature value and the surface temperature measurement value, which conforms to the following equation (23): (23).

[0055] Next, the equation (23) is rearranged to obtain the following equation (24). Unlike Case 1, since the core-to-surface thermal resistance and core temperature value are mainly estimated using surface temperature measurements, the core temperature value of Case 2 will be retained in equation (24). (24).

[0056] The following expression (25) is derived by simplifying expression (24): (25).

[0057] In formula (25), , , and are all model parameters, and can respectively conform to the following formulas (26), (27), (28), (29): (26); (27); (28); (29).

[0058] Similarly, these four model parameters need to be estimated, so that the core to surface thermal resistance can be obtained from the estimated and derived, as shown in the following formula (30): (30).

[0059] In this way, the surface temperature observation operation 320 can model the equation (25) based on the sliding mode observer. In step S03, the surface temperature observation operation 320 observes the battery heat generation, surface temperature measurement value and surrounding temperature measurement value to establish a surface temperature observation model, which conforms to the following equation (31): (31).

[0060] In step S04, the sliding mode operation 330 drives the parameter error equation into the sliding mode, and confirms the parameter error equation into the sliding mode through the Lyapunov function. Similar to Case 1, the control input equation used to update the adaptive control term is as shown in the aforementioned equation (18). In order to satisfy that the first derivative of the Lyapunov function is less than zero, it drives convergence, so the parameter error equation used to update the adaptive parameters is derived, and it conforms to the following equations (32), (33), (34), (35): (32); (33); (34); (35).

[0061] In equations (32), (33), (34), and (35), represents the adaptive gain. In step S05, the adaptive control operation 340, based on the adaptive rule, ensures that the adaptive parameters and adaptive control terms are updated with iterative calculations, causing the adaptive parameters to converge to the model parameters of the true model. Thus, the processing device 300 can derive the estimated parameters corresponding to the convergence of the adaptive parameters to the model parameters from the parameter error equations, where the core-to-surface thermal resistance can be derived from equations (33) and (34), and the core temperature value can be derived from equations (27) and (29), as shown. The remaining parameters can be obtained from the surface temperature measurement and the battery heat generation.

[0062] In step S06, the battery state calculation operation 350 may include a battery thermal resistance calculation operation 351 and a core temperature calculation operation 352. The battery thermal resistance calculation operation 351 includes calculating the core-to-surface thermal resistance of the battery state by performing complex estimated parameters (i.e., AND), that is, the core-to-surface thermal resistance can be derived from the estimated AND, as shown in equation (30). In addition, the core temperature calculation operation 352 includes calculating the core temperature value of the battery state by performing complex estimated parameters (i.e., AND).

[0063] Please refer to Figures 5A, 5B, 6A and 6B, wherein Figure 5A is a graph showing the estimated surface temperature and the measured surface temperature in Case 1 of this disclosure; Figure 5B is a graph showing the percentage error between the estimated surface temperature and the measured surface temperature in Case 1 of this disclosure; Figure 6A is a graph showing the estimated surface temperature and the measured surface temperature in Case 2 of this disclosure; and Figure 6B is a graph showing the percentage error between the estimated surface temperature and the measured surface temperature in Case 2 of this disclosure.

[0064] The adaptive control algorithm disclosed herein can be implemented in Simulink. The Simulink model shows the interconnected components used to simulate battery thermal behavior, including core temperature values, surface temperature measurements, tracking surface-to-ambient thermal resistance, and core-to-surface thermal resistance. The Simulink models in Case 1 and Case 2 illustrate an adaptive control system designed to estimate and manage battery thermal behavior. The Simulink models in both cases aim to dynamically adjust and estimate thermal resistance to ensure effective thermal management. By adjusting adaptive parameters and accurately predicting battery thermal behavior, the adaptive control system maintains a safe operating temperature, thereby improving battery performance and safety.

[0065] The performance of the adaptive control algorithm is evaluated by comparing the estimated temperature curve (i.e., the curve of the surface temperature estimate) and the actual temperature curve (i.e., the curve of the surface temperature measurement). The verification process involves running simulations to observe the degree of matching between the model prediction and the actual thermal behavior of the battery under various operating conditions. As shown in Figures 5A, 5B, 6A, and 6B, the curve of the surface temperature measurement shows how the battery state estimation system 100 and battery state estimation method S0 of this disclosure maintain the battery temperature within a safe range. The temperature initially rises rapidly but quickly stabilizes, demonstrating the effectiveness of the adaptive control algorithm. The initial rapid temperature rise is due to the heat generated exceeding the heat dissipated, and stabilization for approximately 200 seconds indicates that the system has reached thermal equilibrium, where the heat generated equals the heat dissipated. In Case 1, the percentage error between the surface temperature estimate and the surface temperature measurement is very small, at 0.300928%, indicating that the surface temperature observation model of this disclosure accurately reflects the actual situation, and the system reaches a steady state after the initial transient. In Case 2, the surface temperature estimate curve also highlights the ability to manage battery thermal behavior, with the temperature stabilizing after an initial rapid rise. The stability over approximately 200 seconds further indicates that the system has reached thermal equilibrium. The percentage error between the surface temperature estimate and the measured surface temperature is 0.186814%, representing a more accurate surface temperature observation model compared to Case 1.

[0066] Please refer to Figures 7 and 8, where Figure 7 shows the curve of tracking surface-to-ambient thermal resistance in Example 1 of this disclosure, and Figure 8 shows the curve of core-to-surface thermal resistance in Example 2 of this disclosure. The effectiveness of the battery state estimation system 100 and battery state estimation method S0 of this disclosure in estimating thermal resistance can be evaluated through the aforementioned Examples 1 and 2. Examples 1 and 2 focus on different aspects of battery thermodynamics, and the tracking of surface-to-ambient thermal resistance and core-to-surface thermal resistance can be performed by adjusting the adaptive parameters within the adaptive algorithm.

[0067] Case 1 focuses on estimating the surface-to-ambient thermal resistance and tracking it based on the initial conditions of the adaptive parameters, where the initial values ​​of and can be 0.1, 0.001, and 0.01, respectively. These initial values ​​are chosen to reflect the initial conditions of the battery's thermal behavior, but this disclosure is not limited to them. As the simulation progresses, the battery state estimation system 100 dynamically adjusts the adaptive parameters to reflect the instantaneous thermal behavior. As shown in Figure 7, over time, the surface-to-ambient thermal resistance decreases rapidly from a high value and stabilizes at around 3 °C / W after 200 seconds. The rapid decrease in the initial transient behavior indicates a significant initial thermal resistance, possibly due to the initial high current and associated heat generation. Stabilization at 3 °C / W indicates that the system has reached thermal equilibrium, where the surface-to-ambient thermal resistance becomes constant.

[0068] On the other hand, Case 2 focuses on estimating the core-to-surface thermal resistance and tracks it based on the initial conditions of the adaptive parameters, where the initial values ​​of , , and can be 0.1, 0.001, 0.01, and 0.1, respectively, but this disclosure is not limited to these. As shown in Figure 8, within the first 200 seconds, the core-to-surface thermal resistance rapidly decreases from a high value of approximately 10 °C / W and stabilizes at around 0.5 °C / W, indicating that the system has reached thermal equilibrium. Cases 1 and 2 demonstrate the efficiency of the battery state estimation system 100 and battery state estimation method S0 disclosed herein in adapting to changing thermal conditions and maintaining stability.

[0069] In summary, the battery state estimation system and method disclosed herein have the following advantages: First, they can instantly estimate / track the battery's core temperature, core-to-surface thermal resistance, and surface-to-ambient thermal resistance, thus enabling non-destructive measurement to monitor the dynamic state of battery thermal parameters and prevent thermal runaway. Second, they employ adaptive control algorithms to model the battery's thermal behavior, thereby estimating relevant thermal parameters and ensuring the battery operates within its optimal temperature range, thereby improving battery performance, safety, and lifespan, and supporting applications in portable electronic products, electric vehicles, and renewable energy storage. Third, as part of the development options for high-end electric vehicle energy storage systems, they offer advantages in improving battery power density and energy density, and extending lithium-ion battery life.

[0070] Although the present disclosure has been disclosed above with reference to embodiments, it is not intended to limit the present disclosure. Anyone skilled in the art may make various modifications and alterations without departing from the spirit and scope of the present disclosure. Therefore, the scope of protection of the present disclosure shall be determined by the claims attached. [Simplified Explanation of the Diagram]

[0071] Figure 1 is a block diagram illustrating the battery state estimation system of the first embodiment of the present disclosure; Figure 2 is a schematic diagram illustrating the thermal parameters related to the battery in the battery state estimation system of Figure 1; Figure 3 is a flowchart illustrating the battery state estimation method of the second embodiment of the present disclosure; Figure 4 is a flowchart illustrating the various operations in the battery state estimation method of Figure 3; Figure 5A is a graph illustrating the surface temperature estimation value and the surface temperature measurement value in Example 1 of the present disclosure; Figure 5B is a graph illustrating the percentage error between the surface temperature estimation value and the surface temperature measurement value in Example 1 of the present disclosure; Figure 6A is a graph illustrating the surface temperature estimation value and the surface temperature measurement value in Example 2 of the present disclosure; Figure 6B is a graph illustrating the percentage error between the surface temperature estimation value and the surface temperature measurement value in Example 2 of the present disclosure; Figure 7 shows the surface-to-ambient thermal resistance curve in Case 1 of this disclosure; and Figure 8 shows the core-to-surface thermal resistance curve in Case 2 of this disclosure.

Claims

1. A battery state estimation system, comprising: a measuring device for measuring a battery's charge / discharge current, a battery terminal voltage, a surface temperature measurement, and a peripheral temperature measurement; and a processing device connected to the measuring device, and acquiring the battery charge / discharge current, the battery terminal voltage, the surface temperature measurement, and the peripheral temperature measurement, the processing device being configured to perform operations including: performing a heat generation estimation operation on the battery charge / discharge current and the battery terminal voltage to generate battery heat generation; performing a surface temperature observation operation, wherein the surface temperature observation operation includes observing the battery heat generation, the surface temperature measurement, and the peripheral temperature measurement to establish a surface temperature observation model and generate a surface temperature error value, the surface temperature observation model including complex adaptive parameters and an adaptive control term; performing a sliding mode operation on the surface temperature error value to define a sliding function, and defining a complex parameter error equation and a control input equation based on the sliding function; An adaptive control operation is performed, wherein the adaptive control operation updates the adaptive parameters and the adaptive control term through the parameter error equations and the control input equations, respectively, and stabilizes the sliding function within a finite time, so that the parameter error equations converge within the finite time to generate complex estimated parameters; and a battery state calculation operation is performed on the estimated parameters to estimate one of the battery states.

2. The battery state estimation system as described in claim 1, wherein the heat generation estimation operation includes an internal resistance estimation operation and a heat generation calculation operation, the internal resistance estimation operation includes estimating a battery internal resistance value based on the battery charging and discharging current and the battery terminal voltage, and the heat generation calculation operation includes calculating the battery heat generation by combining the battery internal resistance value and the battery charging and discharging current.

3. The battery state estimation system as described in claim 1, wherein the surface temperature observation model further includes the battery heat generation and a surface temperature estimate, and conforms to the following formula: ; where, The estimated surface temperature is given by a first derivative, and the estimated surface temperature is given by a second derivative. , , and are the adaptive parameters, respectively, is the adaptive control term, and is the battery heat generation. The surface temperature observation operation further includes calculating the surface temperature error value by comparing the estimated surface temperature with the measured surface temperature.

4. The battery state estimation system as described in claim 1, wherein the sliding function includes the surface temperature error value and a control gain, and conforms to the following formula: ; where, The sliding function is defined as follows: Let be the surface temperature error value, and let be the control gain after differentiating the surface temperature error value. The sliding mode operation includes driving the parameter error equations into a sliding mode. When the parameter error equations enter the sliding mode, the sliding function reaches a sliding surface, which is 0. The sliding mode operation further includes confirming that the parameter error equations have entered the sliding mode through a Lyapunov function.

5. The battery state estimation system as described in claim 1, wherein the battery state calculation operation includes a battery thermal resistance calculation operation and a core temperature calculation operation, the battery thermal resistance calculation operation includes calculating the estimated parameters to generate a surface-to-ambient thermal resistance or a core-to-surface thermal resistance for the battery state, and the core temperature calculation operation includes calculating the estimated parameters to generate a core temperature value for the battery state.

6. A battery state estimation method, comprising: measuring a battery charging / discharging current, a battery terminal voltage, a surface temperature measurement, and a peripheral temperature measurement of a battery using a measuring device; generating battery heat by performing a heat generation estimation operation on the battery charging / discharging current and the battery terminal voltage using a processing device; performing a surface temperature observation operation using the processing device, wherein the surface temperature observation operation includes observing the battery heat generation, the surface temperature measurement, and the peripheral temperature measurement to establish a surface temperature observation model and generate a surface temperature error value, the surface temperature observation model including complex adaptive parameters and an adaptive control term; defining a sliding function by performing a sliding mode operation on the surface temperature error value using the processing device, and defining a complex parameter error equation and a control input equation based on the sliding function; The processing device performs an adaptive control operation, wherein the adaptive control operation updates the adaptive parameters and the adaptive control terms through the parameter error equations and the control input equations, respectively, and stabilizes the sliding function within a finite time, so that the parameter error equations converge within the finite time to generate complex estimated parameters; and the processing device performs a battery state calculation operation on the estimated parameters to estimate one of the battery states.

7. The battery state estimation method as described in claim 6, wherein the heat generation estimation operation includes an internal resistance estimation operation and a heat generation calculation operation, the internal resistance estimation operation includes estimating a battery internal resistance value based on the battery charging / discharging current and the battery terminal voltage, and the heat generation calculation operation includes calculating the battery heat generation by combining the battery internal resistance value and the battery charging / discharging current.

8. The battery state estimation method as described in claim 6, wherein the surface temperature observation model further includes the battery heat generation and an estimated surface temperature value, and conforms to the following formula: ; where, The estimated surface temperature is given by a first derivative, and the estimated surface temperature is given by a second derivative. , , and are the adaptive parameters, respectively, is the adaptive control term, and is the battery heat generation. The surface temperature observation operation further includes calculating the surface temperature error value by comparing the estimated surface temperature with the measured surface temperature.

9. The battery state estimation method as described in claim 6, wherein the sliding function includes the surface temperature error value and a control gain, and conforms to the following formula: ; where, The sliding function is defined as follows: Let be the surface temperature error value, and let be the control gain after differentiating the surface temperature error value. The sliding mode operation includes driving the parameter error equations into a sliding mode. When the parameter error equations enter the sliding mode, the sliding function reaches a sliding surface, which is 0. The sliding mode operation further includes confirming that the parameter error equations have entered the sliding mode through a Lyapunov function.

10. The battery state estimation method as described in claim 6, wherein the battery state calculation operation includes a battery thermal resistance calculation operation and a core temperature calculation operation, the battery thermal resistance calculation operation includes calculating the estimated parameters to generate a surface-to-ambient thermal resistance or a core-to-surface thermal resistance of the battery state, and the core temperature calculation operation includes calculating the estimated parameters to generate a core temperature value of the battery state.

Images