A method and system for estimating internal temperature of a lithium-ion battery

Abstract

This invention provides a method and system for estimating the internal temperature of a lithium-ion battery, relating to the field of battery management technology. The method includes: an offline calibration stage and an online application stage; collecting multi-condition experimental data of the lithium-ion battery; identifying electrical model parameters based on the multi-condition experimental data; identifying thermal model parameters based on the multi-condition experimental data and the electrical model parameters; constructing an open-circuit strain-state-of-charge mapping function; calculating parameter sample points based on the thermal model parameters; establishing a parametric function model based on the parameter sample points; deploying the parametric function model and the open-circuit strain-state-of-charge mapping function to the battery management system; collecting real-time state data of the lithium-ion battery; and estimating the internal temperature of the lithium-ion battery in real time through the battery management system based on the real-time state data.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, and in particular to a method and system for estimating the internal temperature of a lithium-ion battery. Background Technology

[0002] With the rapid development of new energy storage, electric vehicles, and portable electronic devices, lithium-ion batteries have become the core carrier in the current energy storage and power supply fields due to their advantages such as high energy density, long cycle life, and excellent charge and discharge performance. During actual charge and discharge operation, lithium-ion batteries undergo complex electrochemical reactions, lithium-ion intercalation / deintercalation volume deformation, and heat generation and dissipation. Internal temperature is a key parameter directly affecting battery safety, cycle life, charge and discharge efficiency, and the risk of thermal runaway. Accurately obtaining the real-time internal temperature of the battery is of vital practical significance for battery thermal management and control, fault warning, safety protection, and full life-cycle health management.

[0003] Currently, most lithium-ion battery temperature detection in the industry adopts traditional contact temperature measurement methods, which mainly collect external temperature by attaching temperature sensors to the battery surface. At the same time, some solutions rely on the battery equivalent circuit model and thermal model combined with the ampere-hour integration method to estimate the battery state, or use finite element simulation and multiphysics modeling to simulate the internal temperature distribution of the battery, thereby indirectly characterizing the internal thermal state of the battery. This has become the mainstream technical means for battery temperature sensing and state assessment.

[0004] However, existing lithium-ion battery temperature measurement technologies have significant shortcomings: internal sensors require sealing holes in the battery casing, which damages the original battery structure and introduces safety hazards and increased manufacturing complexity; while relying solely on surface temperature measurement is affected by the lag in heat conduction, making it difficult to reflect the true internal thermal state during the battery's heating phase, resulting in delayed warnings. Furthermore, traditional modeling and estimation methods are computationally complex and require high hardware computing power, making them unsuitable for real-time operation in embedded battery management system scenarios. Summary of the Invention

[0005] To address the significant shortcomings of existing lithium-ion battery temperature measurement technologies, which rely on surface temperature measurement to fail to reflect the true internal core temperature, resulting in large temperature differences between the internal and external components and low measurement accuracy, and to overcome the technical challenges of traditional modeling and estimation methods—namely, their high computational complexity, high hardware computing power requirements, and difficulty in real-time operation within embedded battery management system scenarios—this technology aims to address these issues.

[0006] The technical solution provided by this invention is as follows: The first aspect of this invention provides a method for estimating the internal temperature of a lithium-ion battery, comprising: an offline calibration stage and an online application stage; The offline calibration phase specifically includes steps S1 to S7; S1: Collect multi-condition experimental data of lithium-ion batteries; S2: Identify the electrical model parameters based on multi-condition experimental data; S3: Identify the thermal model parameters based on multi-condition experimental data and electrical model parameters; S4: Construct the open-circuit strain-charge state mapping function; S5: Calculate parameter sample points based on the thermal model parameters; S6: Establish a parametric function model based on the parameter sample points; S7: Deploy the parametric function model and open-circuit strain-state-of-charge mapping function to the battery management system; The online application phase specifically includes steps S8 to S9; S8: Collects real-time status data of lithium-ion batteries; S9: Based on real-time status data, the internal temperature of the lithium-ion battery is estimated in real time through the battery management system.

[0007] A second aspect of the present invention provides a lithium-ion battery internal temperature estimation system, comprising: processor; The memory stores computer-readable instructions, which, when executed by the processor, implement the lithium-ion battery internal temperature estimation method as described in the first aspect.

[0008] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the lithium-ion battery internal temperature estimation method of the first aspect.

[0009] The beneficial effects of the technical solution provided by this invention include: In this embodiment of the invention, the internal temperature of the lithium-ion battery is estimated in real time by the battery management system, eliminating the need for sealing holes in the battery casing. This avoids damaging the original battery structure, reduces safety hazards, and decreases process complexity. By deploying the parametric function model and the open-circuit strain-state-of-charge mapping function into the battery management system, the thermal conduction hysteresis problem caused by relying on surface temperature measurement can be avoided, allowing for timely reflection of the true internal thermal state during the battery heating stage. Furthermore, since the two types of function models are pre-deployed into the battery management system, online applications do not require complex simulations or large-scale computations. The overall computational logic is simplified, with low computational load, making it well-suited for the low-computing-power embedded operation scenarios of the battery management system. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a method for estimating the internal temperature of a lithium-ion battery according to an embodiment of the present invention. Figure 2This is a schematic diagram of a sensor arrangement provided in an embodiment of the present invention.

[0011] Figure 3 This is a schematic diagram of the internal temperature estimation system of a lithium-ion battery provided in an embodiment of the present invention. Detailed Implementation

[0012] Reference manual attached Figure 1 The diagram shows a flowchart of a method for estimating the internal temperature of a lithium-ion battery according to an embodiment of the present invention.

[0013] This invention provides a method for estimating the internal temperature of a lithium-ion battery, comprising: an offline calibration stage and an online application stage.

[0014] The offline calibration stage specifically includes steps S1 to S7.

[0015] S1: Collect multi-condition experimental data of lithium-ion batteries.

[0016] Optionally, the multi-condition experimental data specifically include: external strain, external temperature, internal temperature, ambient temperature, current, voltage, and state of charge.

[0017] S2: Identify the electrical model parameters based on multi-condition experimental data.

[0018] In one possible implementation, S2 specifically involves: identifying the electrical model parameters based on multi-condition experimental data using a first-order RC equivalent circuit model and a forgetting factor recursive least squares algorithm.

[0019] Among them, the first-order RC equivalent circuit model is a commonly used lumped parameter model for lithium-ion batteries. It equates the dynamic characteristics of the battery to an ohmic internal resistance, a first-order RC circuit consisting of polarization resistance and polarization capacitance, and an open-circuit voltage source that changes with the state of charge. It can simulate the ohmic voltage drop and polarization voltage drop of the battery during charging and discharging, providing a basic circuit characterization for subsequent parameter identification, heat generation calculation and state estimation.

[0020] Among them, the Forgotten Factor Recursive Least Squares (FFRLS) algorithm is a recursive algorithm for online parameter identification of time-varying systems. By introducing a forgetting factor, it weakens the weight of historical data and strengthens the influence of new data, avoiding the data saturation problem of traditional recursive least squares. It can track the dynamic changes of parameters in the first-order RC equivalent circuit model with temperature, aging, and state of charge in real time, providing accurate dynamic parameters for subsequent heat generation rate calculation and internal temperature estimation.

[0021] The first-order RC equivalent circuit model is as follows: in, Up Indicates polarization voltage. t Indicates time, R p Indicates polarization resistance, C p Indicates polarization capacitance. I Indicates current (positive for discharge). U OCV represents the open-circuit voltage, indicating the terminal voltage. SOC Indicates the state of charge. R 0 represents ohmic internal resistance.

[0022] Specifically, in the offline calibration stage, using the experimentally measured voltage and current data, the recursive least squares method with a forgetting factor (FFRLS) is used to identify R0 and R2 online. p C p This allows us to obtain parameter values ​​under different operating conditions.

[0023] It should be noted that the electrical model is used to estimate the state of charge (SOC) in real time (using the ampere-hour integral method), and also provides ohmic resistance and polarization parameters for subsequent heat generation calculations in the thermal model.

[0024] In this embodiment of the invention, by identifying electrical model parameters based on a first-order RC equivalent circuit model and a forgetting factor recursive least squares algorithm, accurate dynamic tracking of the ohmic internal resistance, polarization resistance, and polarization capacitance of lithium-ion batteries can be achieved. This not only solves the data saturation problem of traditional algorithms but also captures the time-varying characteristics of parameters with temperature, aging, and state of charge.

[0025] S3: Identify the thermal model parameters based on multi-condition experimental data and electrical model parameters.

[0026] It should be noted that, due to the relatively small temperature gradient inside a cylindrical battery, a lumped-parameter thermal model can be used to describe the battery's thermal behavior. This model treats the battery as a uniformly heated body and is based on the law of energy conservation.

[0027] In one possible implementation, S3 specifically involves: identifying thermal model parameters based on multi-condition experimental data and electrical model parameters using a lumped parameter thermal model.

[0028] Among them, the lumped parameter thermal model is a lumped thermal characterization method that simplifies lithium-ion batteries to a single or a few temperature nodes. It ignores the internal temperature gradient of the battery, constructs a lumped thermal balance equation through equivalent thermal resistance, thermal capacity and heat generation rate parameters, calculates the heat generation during the charging and discharging process of the battery by combining ohmic internal resistance, polarization internal resistance and entropy heat effect, and describes heat transfer through heat conduction and convection heat dissipation terms. It can quickly solve the average temperature or core temperature of the battery with low computational complexity, and provide reliable thermal data support for obtaining internal temperature reference values ​​in the offline calibration stage and then identifying mechanical model parameters.

[0029] The lumped parameter thermal model is specifically as follows: in, m Indicates battery quality. c p Indicates specific heat capacity. T b This indicates the battery temperature (equal to the internal temperature T). in ), q h Indicates the total heat production rate. q s This indicates the total heat dissipation rate.

[0030] Among them, total heat generation rate q h It consists of three parts: in, q r Indicates the heat of reaction. Represents the entropy-heat coefficient. q J Indicates Joule heating. q D Indicates polarization heat, I p This represents the current flowing through the polarization resistor. T a Indicates ambient temperature. R T Indicates thermal resistance.

[0031] In this embodiment of the invention, based on multi-condition experimental data and identified electrical model parameters, a lumped-parameter thermal model is used to identify thermal model parameters. The small internal temperature gradient of a cylindrical battery is utilized for lumped simplification, eliminating the need to construct a complex multidimensional temperature field and significantly reducing computational complexity. Simultaneously, reaction heat, Joule heat, and polarization heat are combined to accurately characterize the battery's heat generation mechanism, and the heat dissipation process is quantified through thermal resistance, which is used to obtain internal temperature reference values ​​or for thermal parameter identification.

[0032] S4: Construct the open-circuit strain-charge state mapping function.

[0033] Among them, the open-circuit strain-state-of-charge mapping function is a functional model that characterizes the one-to-one correspondence between the open-circuit strain of the battery shell and the state of charge of a lithium-ion battery under static equilibrium state. It is obtained by collecting and fitting the surface strain data of the battery after static stabilization under different states of charge and different temperature conditions. It can reflect the static strain performance of the battery shell caused by the volume change caused by lithium ion insertion and extraction. It provides a key basis for removing the static strain component corresponding to the state of charge from the measured total strain and extracting the dynamic strain signal caused only by temperature and dynamic process.

[0034] It should be noted that open-circuit strain (OCS) is defined as the strain value of a battery in a fully balanced state, which is only related to the state of charge (SOC). An open-circuit strain-state of charge (OCS-SOC) mapping is established by directly measuring the equilibrium strain at different SOC points through OCS experiments.

[0035] Specifically, the open-circuit strain-charge state mapping function is as follows: in, OCS Indicates open-circuit strain, focs The mapping function representing open-circuit strain-charged state.

[0036] In this embodiment of the invention, an open-circuit strain-state-of-charge mapping function is constructed. The measured strain data under the static equilibrium condition of the battery is used to establish an accurate correspondence between the two. This can effectively remove the static strain component caused by the change in state of charge, accurately separate the dynamic strain that is not affected by the state of charge from the total strain on the battery surface, eliminate the influence of the battery lithium insertion / extraction volume deformation on stress and strain detection, and lay a pure and reliable strain data foundation for the subsequent establishment of thermo-mechanical coupling relationship and accurate solution of internal temperature change, which greatly improves the accuracy and stability of battery internal temperature estimation.

[0037] S5: Calculate parameter sample points based on the thermal model parameters.

[0038] In one possible implementation, S5 specifically includes sub-steps S501 to S503: S501: Corrects the internal temperature change based on the thermal model parameters.

[0039] S502: Calculate the change in state of charge and dynamic strain.

[0040] Dynamic strain refers to the strain signal remaining after removing the static open-circuit strain component caused by changes in the state of charge from the total strain measured on the battery surface during the charging and discharging process of a lithium-ion battery. This strain is generated by the combined effects of temperature changes, electrochemical polarization, and dynamic charging and discharging processes, and directly reflects the changes in the internal thermal state and dynamic reaction processes of the battery.

[0041] The specific formula for calculating dynamic strain is as follows: in, t Indicates time, S p ( t )express t Dynamic response at all times express t External strain at any moment OCS ( t )express t Always be ready to open up new avenues and respond to emergencies.

[0042] Among them, external strain It includes equilibrium components and dynamic components.

[0043] S503: Based on the thermo-mechanical coupling relationship, the internal temperature change, charge state change and dynamic strain are identified, and parameter sample points are calculated.

[0044] Among them, the thermo-mechanical coupling relationship is a mathematical model describing the linear relationship between the dynamic strain of the battery shell and the two types of changes caused by the changes in state of charge and internal temperature during the charging and discharging process of lithium-ion batteries. It decomposes the dynamic strain into the electro-induced strain component caused by the change in state of charge and the thermo-induced strain component caused by the temperature change, and introduces the dynamic state of charge strain coefficient and the dynamic temperature strain coefficient that change dynamically with the operating conditions, respectively. It constructs a mapping relationship from the externally measurable dynamic strain and the change in state of charge to the internal temperature change, and provides the core theoretical basis for the subsequent real-time recursive estimation of the battery internal temperature.

[0045] The specific formula for calculating the thermo-mechanical coupling relationship is as follows: in, Indicates the sampling time. S p ( ) indicates the first Dynamic strain at each sampling time, a SOC ( ) indicates the first The dynamic state-of-charge strain coefficient at each sampling time, ∆ SOC ( ) indicates the first The change in state of charge at each sampling time, α T ( ) indicates the first The dynamic temperature strain coefficient at each sampling time, ∆ Tin ( ) indicates the first The change in internal temperature of the lithium-ion battery at each sampling time.

[0046] It should be noted that by transforming the formula for the thermo-mechanical coupling relationship, the formula for the internal temperature change can be obtained: Furthermore, the current internal temperature can be recursively derived from the temperature at the previous moment: in, T in ( ) indicates the first The internal temperature of the lithium-ion battery at each sampling time. T in ( -1) indicates the first -1 sampling time of the internal temperature of the lithium-ion battery, ∆ SOC ( -1) indicates the first -1 change in state of charge at sampling time.

[0047] The parameter sample points include the operating condition input variables and the corresponding strain coefficient output variables. The operating condition input variables include ambient temperature, external temperature, external strain, and state of charge. The strain coefficient output variables include the dynamic state of charge strain coefficient and the dynamic temperature strain coefficient.

[0048] In this embodiment of the invention, by establishing a linear coupling relationship between the dynamic strain measured outside the battery and the changes in state of charge and internal temperature, a precise mapping from measurable operating condition variables (ambient temperature, external temperature, external strain, SOC) to key parameters (dynamic state of charge strain coefficient and dynamic temperature strain coefficient) is achieved. This generates parameter sample points during the offline calibration stage, providing a reliable basis for subsequent online recursive calculation of internal temperature.

[0049] S6: Establish a parametric function model based on the parameter sample points.

[0050] Among them, the parametric function model refers to a nonlinear mapping function model obtained by fitting and training the dynamic state-of-charge strain coefficient and dynamic temperature strain coefficient samples obtained in the offline calibration stage with ambient temperature, external temperature, external strain and state of charge as independent variables through methods such as polynomial regression, neural network or support vector regression. This model can characterize the continuous change law of dynamic strain coefficient with operating conditions, and provide key dynamic parameter support for real-time calculation of dynamic strain coefficient at the current moment in the online application stage, and then recursively solve the internal temperature of the battery.

[0051] Specifically, regression analysis is performed on the obtained parameter sample points and the corresponding operating condition variables to establish a parametric function model. Three methods can be used for fitting and modeling: multinomial regression, neural network, and support vector regression.

[0052] Specifically, multinomial regression is as follows: in, a SOC Indicates the strain coefficient under dynamic charging state. i This indicates the order of the polynomial corresponding to the ambient temperature. j This indicates the order of the polynomial corresponding to the external temperature. m This indicates the order of the polynomial corresponding to the external strain. l This indicates the order of the polynomial corresponding to the state of charge. n This indicates the highest order of the polynomial fit. β ijml This represents the polynomial regression coefficients corresponding to the strain coefficients under dynamic charging state. Indicates ambient temperature i Power term, Indicates external temperature j Power term, Indicating external strain m Power term, SOC l Indicates the state of charge l Power term, α T Represents the dynamic temperature strain coefficient. γ ijml This represents the polynomial regression coefficient corresponding to the dynamic temperature strain coefficient.

[0053] Among them, neural networks are ( T amb , T ext , , SOC ) is the input, ( a SOC , α T ) is the output, used to train the feedforward neural network.

[0054] in, T amb Indicates ambient temperature. T ext Indicates the external temperature. Indicates external strain. SOC Indicates the state of charge.

[0055] Furthermore, the fitted function is: in, g and h Both represent nonlinear mapping functions.

[0056] In this embodiment of the invention, based on the acquired parameter sample points, multiple modeling methods such as multinomial regression, neural networks, and support vector regression are employed. Using ambient temperature, external temperature, external strain, and state of charge as input variables, a nonlinear parametric function model of dynamic state of charge strain coefficient and dynamic temperature strain coefficient is fitted and constructed. This model accurately depicts the nonlinear variation of these two types of strain coefficients with multiple operating conditions, overcoming the limitation that fixed coefficients cannot adapt to complex operating conditions. Simultaneously, multiple optional modeling paths are provided to adapt to the accuracy and computing power requirements of different application scenarios. Discrete parameter samples are transformed into continuous mapping relationships that can be called in real time, providing stable and reliable model support for subsequent online monitoring to quickly solve for strain coefficients time-by-time and recursively estimate the internal temperature of the battery with high accuracy.

[0057] S7: Deploy the parametric function model and open-circuit strain-state-of-charge mapping function to the battery management system.

[0058] Among them, the battery management system is an embedded control and protection unit for lithium-ion battery packs. It collects state data such as voltage, current, temperature and strain of individual battery cells, estimates key parameters such as state of charge, health status and internal temperature in real time, and executes functions such as charge and discharge control, equalization management, thermal safety warning and fault diagnosis according to preset strategies. It provides safe, stable and efficient operation guarantee for the battery and is the core control center for reliable operation and safe operation of the battery system.

[0059] In this embodiment of the invention, the parametric function model constructed in the offline stage and the open-circuit strain-state-of-charge mapping function are uniformly deployed and integrated into the battery management system. This enables the battery management system to have built-in model capabilities for real-time solution of operating parameters and automatic static strain stripping, eliminating the need for additional offline computing equipment. The model can be locally invoked by relying on the system's embedded computing power. Simultaneously, it achieves seamless conversion of offline calibration results into online engineering applications, providing comprehensive model support for subsequent real-time acquisition of operating data, online dynamic calculation of strain coefficients, accurate decoupling of strain components, and recursive estimation of battery internal temperature. This ensures that the entire detection and temperature estimation process can operate autonomously, continuously, and with low latency in a closed loop.

[0060] The online application phase specifically includes steps S8 to S9.

[0061] S8: Collects real-time status data of lithium-ion batteries.

[0062] It should be noted that during the online application phase, the system can recursively estimate the internal temperature by collecting external strain, external temperature, ambient temperature, and state of charge (provided in real time by the battery management system based on the equivalent circuit model), thus eliminating the need for an internal temperature sensor.

[0063] Furthermore, the initial external temperature was read. T ext (0), let .

[0064] Reference manual attached Figure 2 The diagram shows a sensor arrangement according to an embodiment of the present invention.

[0065] It should be noted that a strain sensor and an external temperature sensor are attached to the axial center of the cylindrical lithium-ion battery casing, while the ambient temperature sensor is placed in the air near the battery.

[0066] in, T in Indicates internal temperature. T ext Indicates the external temperature. ε Indicates external strain.

[0067] S9: Based on real-time status data, the internal temperature of the lithium-ion battery is estimated in real time through the battery management system.

[0068] In one possible implementation, S9 specifically includes sub-steps S901 to S904: S901: Obtain the current state of charge through the battery management system.

[0069] S902: Calculate the current dynamic strain based on the current state of charge using the open-circuit strain-state of charge mapping function.

[0070] In one possible implementation, S902 specifically includes steps S9021 and S9022: S9021: Calculate the current open-circuit strain based on the current state of charge using the open-circuit strain-state of charge mapping function.

[0071] S9022: Calculate the current dynamic strain based on real-time status data and current open-circuit strain.

[0072] The specific formula for calculating the current dynamic strain is as follows: in, k Indicates the sampling time. S p ( k ) indicates the firstk Dynamic strain at each sampling time, Indicates the first k External strain at each sampling time, OCS ( k ) indicates the first k Open circuit strain at each sampling time.

[0073] S903: Based on real-time status data, calculate the current dynamic state of charge strain coefficient and the current dynamic temperature strain coefficient through a parametric function model.

[0074] Among them, the dynamic state of charge strain coefficient a SOC It reflects the change in electrode volume caused by lithium-ion insertion / extraction, which is related to materials, structure, and aging, and is also affected by ambient temperature, external temperature, strain, and state of charge.

[0075] Among them, dynamic temperature strain coefficient α T It reflects the thermal expansion characteristics of battery materials, which are related to temperature, state of charge, and aging, and are also affected by ambient temperature, external temperature, strain, and state of charge.

[0076] S904: Estimate the internal temperature of the lithium-ion battery in real time based on the current dynamic strain, the current dynamic state of charge strain coefficient, and the current dynamic temperature strain coefficient.

[0077] In one possible implementation, S904 specifically includes steps S9041 to S9043: S9041: Calculate the change in the current state of charge based on the current state of charge.

[0078] The specific formula for calculating the change in the current state of charge is as follows: Where, ∆ SOC ( k ) indicates the first k The change in state of charge at each sampling time, SOC ( k ) indicates the first k State of charge at each sampling time, SOC ( k -1) indicates the first k -1 state of charge at sampling time.

[0079] S9042: Calculate the internal temperature change of the lithium-ion battery based on the current change in state of charge, the current dynamic strain, the current dynamic state of charge strain coefficient, and the current dynamic temperature strain coefficient.

[0080] The specific formula for calculating the internal temperature change is as follows: Where, ∆ T in ( k ) indicates the first k The internal temperature change at each sampling time, S p ( k ) indicates the first k Dynamic strain at each sampling time, a SOC ( k ) indicates the first k The dynamic state-of-charge strain coefficient at each sampling time, α T ( k ) indicates the first k The dynamic temperature strain coefficient at each sampling time.

[0081] S9043: Estimates the internal temperature of a lithium-ion battery in real time based on changes in internal temperature.

[0082] The specific formula for calculating the internal temperature is as follows: in, T in ( k ) indicates the first k Internal temperature at each sampling time, T in ( k -1) indicates the first k -1 internal temperature at sampling time.

[0083] Furthermore, the internal temperature is output to the battery management system for thermal management, safety warnings, and other purposes.

[0084] In this embodiment of the invention, various state data are collected in real time by the battery management system. The open-circuit strain is accurately solved and the dynamic strain is separated by the open-circuit strain-state-of-charge mapping function. Then, the dynamic state-of-charge strain coefficient and the dynamic temperature strain coefficient are calculated in real time by calling the deployed parameter function model. The change in state of charge is solved by combining the difference in state of charge at different times. Then, the internal temperature change is calculated step by step and the real-time internal temperature of the battery is obtained by time-series recursion. The entire process does not require the installation of temperature sensors inside the battery. Non-invasive continuous temperature estimation can be achieved by relying only on external measurable operating conditions and strain data. The calculation logic is simple, the real-time performance is strong, and it is adaptable to complex and ever-changing actual operating conditions. The estimation results have high accuracy and good stability, which can provide continuous and reliable internal temperature data support for battery thermal management and control, early warning of thermal faults, and safe operation control.

[0085] Reference manual attached Figure 3 The diagram shows a schematic of the internal temperature estimation system for a lithium-ion battery provided in an embodiment of the present invention.

[0086] The present invention also provides a lithium-ion battery internal temperature estimation system 20, applied to the above-mentioned lithium-ion battery internal temperature estimation method, comprising: The strain sensing unit is used to collect external strain of the battery.

[0087] Optionally, the strain sensing unit specifically includes: a fiber Bragg grating sensor and multiple FBG measuring points arranged at intervals along the battery axis.

[0088] A fiber Bragg grating sensor is attached to the surface of the battery casing along the axial direction of a cylindrical lithium-ion battery to collect axial external strain.

[0089] FBG measuring points are used to obtain the axial strain distribution of the battery.

[0090] An external temperature sensing unit is used to collect the temperature of the battery's outer surface.

[0091] An ambient temperature sensing unit is used to collect ambient temperature data.

[0092] The state of charge acquisition unit is used to acquire the state of charge.

[0093] The storage unit is used to store the open-circuit strain-state-of-charge mapping function, the dynamic state-of-charge strain coefficient function, and the dynamic temperature strain coefficient function.

[0094] The processor is used to recursively estimate the internal temperature based on the above data.

[0095] The output unit is used to output the internal temperature to the battery management system.

[0096] This invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the lithium-ion battery internal temperature estimation method as described in the method embodiment.

[0097] The present invention provides a computer-readable storage medium that can implement the steps and effects of the lithium-ion battery internal temperature estimation method of the above-described method embodiments. To avoid repetition, the present invention will not repeat them.

[0098] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for estimating the internal temperature of a lithium-ion battery, characterized in that, include: Offline calibration stage and online application stage; The offline calibration stage specifically includes steps S1 to S7; S1: Collect multi-condition experimental data of lithium-ion batteries; S2: Identify the electrical model parameters based on the multi-condition experimental data; S3: Identify the thermal model parameters based on the multi-condition experimental data and the electrical model parameters; S4: Construct the open-circuit strain-charge state mapping function; S5: Calculate parameter sample points based on the thermal model parameters; S6: Establish a parametric function model based on the parameter sample points; S7: Deploy the parametric function model and the open-circuit strain-state-of-charge mapping function to the battery management system; The online application phase specifically includes steps S8 to S9; S8: Collect real-time status data of the lithium-ion battery; S9: Based on the real-time status data, the internal temperature of the lithium-ion battery is estimated in real time through the battery management system.

2. The method for estimating the internal temperature of a lithium-ion battery according to claim 1, characterized in that: The multi-condition experimental data specifically include: external strain, external temperature, internal temperature, ambient temperature, current, voltage, and state of charge.

3. The method for estimating the internal temperature of a lithium-ion battery according to claim 1, characterized in that: Specifically, S2 is: Based on the multi-condition experimental data, the electrical model parameters are identified using a first-order RC equivalent circuit model and a forgetting factor recursive least squares algorithm.

4. The method for estimating the internal temperature of a lithium-ion battery according to claim 1, characterized in that: Specifically, S3 is: Based on the multi-condition experimental data and the electrical model parameters, the thermal model parameters are identified using the lumped parameter thermal model.

5. The method for estimating the internal temperature of a lithium-ion battery according to claim 1, characterized in that: S5 specifically includes: S501: Correct the internal temperature change based on the thermal model parameters; S502: Calculate the change in state of charge and dynamic strain; S503: Based on the thermo-mechanical coupling relationship, the internal temperature change, the charge state change, and the dynamic strain are identified, and parameter sample points are calculated.

6. The method for estimating the internal temperature of a lithium-ion battery according to claim 1, characterized in that: S9 specifically includes: S901: Obtain the current state of charge through the battery management system; S902: Based on the current state of charge, calculate the current dynamic strain using the open-circuit strain-state of charge mapping function; S903: Based on the real-time status data, calculate the current dynamic state of charge strain coefficient and the current dynamic temperature strain coefficient using the parameter function model; S904: Estimate the internal temperature of the lithium-ion battery in real time based on the current dynamic strain, the current dynamic state of charge strain coefficient, and the current dynamic temperature strain coefficient.

7. The method for estimating the internal temperature of a lithium-ion battery according to claim 6, characterized in that: Specifically, S902 includes: S9021: Based on the current state of charge, calculate the current open-circuit strain using the open-circuit strain-state of charge mapping function; S9022: Calculate the current dynamic strain based on the real-time status data and the current open-circuit strain.

8. The method for estimating the internal temperature of a lithium-ion battery according to claim 6, characterized in that: Specifically, S904 includes: S9041: Calculate the change in the current state of charge based on the current state of charge; S9042: Calculate the internal temperature change of the lithium-ion battery based on the current state of charge change, the current dynamic strain, the current dynamic state of charge strain coefficient, and the current dynamic temperature strain coefficient; S9043: Estimate the internal temperature of the lithium-ion battery in real time based on the internal temperature change.

9. A lithium-ion battery internal temperature estimation system, applied to the lithium-ion battery internal temperature estimation method according to any one of claims 1 to 8, characterized in that, include: The strain sensing unit is used to collect the external strain of the battery. An external temperature sensing unit is used to collect the temperature of the battery's outer surface. An ambient temperature sensing unit is used to collect ambient temperature data. A state of charge acquisition unit is used to acquire the state of charge. The storage unit is used to store the open-circuit strain-state of charge mapping function, the dynamic state of charge strain coefficient function, and the dynamic temperature strain coefficient function. The processor is configured to recursively estimate the internal temperature based on the open-circuit strain-state-of-charge mapping function, the dynamic state-of-charge strain coefficient function, and the dynamic temperature strain coefficient function. The output unit is used to output the internal temperature to the battery management system.

10. The lithium-ion battery internal temperature estimation system according to claim 9, characterized in that, The strain sensing unit specifically includes: a fiber Bragg grating sensor and multiple FBG measuring points arranged at intervals along the battery axis; The fiber Bragg grating sensor is attached to the surface of the battery casing along the axial direction of the cylindrical lithium-ion battery to collect axial external strain. The FBG measuring points are used to obtain the axial strain distribution of the battery.

Images