A lithium battery internal temperature and heating power online detection system and method based on heat flow and surface temperature

Abstract

The application discloses a kind of lithium battery internal temperature and heat generation power online detection system and method based on heat flow and surface temperature, method includes: step 1: with the flexible heat flow sensor layer of high-precision surface temperature sensor attached is placed in isothermal environment cabin;Step 2: control isothermal environment cabin temperature to set value;Step 3: lithium battery is applied electric load;Step 4: real-time synchronous acquisition heat flux density signal and temperature signal on the surface of lithium battery;Step 5: according to heat flux density signal, the instantaneous total heat generation power of battery is calculated;Step 6: combined with temperature signal, ambient temperature and instantaneous total heat generation power, based on heat model inversion, the internal temperature of battery is calculated.The application directly measures the heat flux density and temperature on the surface of battery, combined with heat conduction model, and the instantaneous heat generation power of battery is calculated in real time, high-precision, non-invasively and the internal temperature thereof is inverted, suitable for lithium battery thermal behavior analysis, state evaluation and safety warning.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of lithium battery thermal management test, and particularly relates to a lithium battery internal temperature and heating power online detection system and method based on heat flow and surface temperature. BACKGROUND

[0002] The thermal safety of a lithium battery is a key restrictive factor for its wide application in the fields of electric vehicles, energy storage systems and the like. The heating power and internal temperature of the battery are core parameters for evaluating the thermal behavior, state (SOH, SOF) and safety risk (such as thermal runaway) of the battery.

[0003] At present, the main equipment for measuring the thermal characteristics of a battery is an isothermal calorimeter (such as ARC). The basic principle is to place the battery in an adiabatic or isothermal environment, and indirectly calculate the heat generation by measuring the energy required to maintain the environmental temperature constant. However, the traditional method has the following limitations:

[0004] 1. Indirect measurement and thermal inertia delay, difficult to capture transient thermal changes;

[0005] 2. Unable to distinguish local heat sources, only the overall heat generation can be obtained;

[0006] 3. Difficult to estimate the internal temperature, the implanted sensor will damage the structure, and the surface temperature estimation model has limited precision.

[0007] Therefore, there is an urgent need for a method and device that can directly, quickly and non-invasively measure the heating power of the battery and accurately calculate the internal temperature. SUMMARY

[0008] To solve the problems of thermal inertia delay, inability to locate local heat sources and low internal temperature estimation accuracy of the traditional isothermal calorimeter, the application provides a lithium battery internal temperature and heating power online detection system and method based on heat flow and surface temperature. The system integrates a heat flow sensor and a surface temperature sensor, directly measures the heat flow density and temperature of the battery surface, combines a heat conduction model, and realizes real-time, high-precision and non-invasive calculation of the instantaneous heating power of the battery and inversion of the internal temperature of the battery, and is suitable for lithium battery thermal behavior analysis, state evaluation and safety warning.

[0009] To achieve the above purpose, the application provides the following scheme:

[0010] A lithium battery internal temperature and heating power online detection system based on heat flow and surface temperature, the system comprising: an isothermal environment cabin, a battery clamp, at least one heat flow sensing module, a data acquisition card, a main control system, a heating / cooling unit, a lithium battery, a high-precision surface temperature sensor, a flexible heat flow sensor layer, an electrical load module and a flexible heat-conducting gasket.

[0011] The isothermal environment cabin is used for providing a constant temperature environment, and the inner wall is provided with a heating / cooling unit, which is controlled by a master control system to maintain a set temperature.

[0012] The battery clamp is used for fixing the lithium battery.

[0013] The lithium battery is a test object, and a high-precision surface temperature sensor and a flexible heat flow sensor layer are attached to the surface, respectively.

[0014] The high-precision surface temperature sensor adopts a thin film platinum resistance or an ultrathin thermocouple structure and is used for real-time measurement of the surface temperature of the battery.

[0015] The flexible heat flow sensor layer is a thermopile array based on the Seebeck effect, which is used for converting the heat flow density vertically through the surface of the battery into an electrical signal output.

[0016] The electrical load module is connected to the lithium battery through a wire and is used for applying a charging and discharging current in the experiment process to simulate the battery heating behavior under different working conditions.

[0017] The flexible heat-conducting gasket is arranged between the surface of the battery and the sensor layer and is used for filling the micro gap.

[0018] The data acquisition card is used for real-time synchronous acquisition of the heat flow density and temperature signals.

[0019] The master control system is used for calculating the instantaneous total heating power of the battery according to the heat flow density signal; and the internal temperature of the battery is calculated based on a thermal model inversion in combination with the temperature signal, the environmental temperature and the instantaneous total heating power.

[0020] The application also provides a lithium battery internal temperature and heating power online detection method based on heat flow and surface temperature.

[0021] Step 1: the lithium battery attached with a high-precision surface temperature sensor and a flexible heat flow sensor layer is placed in the isothermal environment cabin.

[0022] Step 2: the temperature of the isothermal environment cabin is controlled to be stable to a set value.

[0023] Step 3: an electrical load is applied to the lithium battery.

[0024] Step 4: the heat flow density signal and the temperature signal of the surface of the lithium battery are synchronously acquired in real time.

[0025] Step 5: the instantaneous total heating power of the battery is calculated according to the heat flow density signal.

[0026] Step 6: the internal temperature of the battery is calculated based on a thermal model inversion in combination with the temperature signal, the environmental temperature and the instantaneous total heating power.

[0027] Preferably, the method for calculating the instantaneous total heat generation power of the battery according to the heat flux signal is:

[0028] ;

[0029] wherein, is the heat flux density measured by the heat flux sensor; A is the effective heat dissipation area of the battery in contact with the sensor.

[0030] Preferably, the method for constructing the thermal model is:

[0031] According to the actual size of the lithium battery to be measured, a three-dimensional solid geometric model thereof is created, and the battery geometry is discretized into a plurality of volume elements, i.e. mesh elements, using a finite element analysis method;

[0032] Material thermal physical properties are assigned to all mesh elements;

[0033] On each mesh element, a control equation, i.e. a transient heat conduction equation, is set;

[0034] On each mesh element, boundary conditions and initial conditions and source term loading are set, and finally the thermal model is constructed.

[0035] Preferably, the transient heat conduction equation is:

[0036] ;

[0037] wherein, is the density of the battery material; is the specific heat capacity of the battery material; T is the temperature; t is the time; k is the thermal conductivity of the battery material; is the internal heat source generation rate, is the gradient operator.

[0038] Preferably, the method for calculating the internal temperature of the battery based on the thermal model includes:

[0039] The control equation is weighted residual approximated on each mesh element using a finite element analysis method, and the continuous partial differential equation is converted into a system of ordinary differential equations about the temperature of all nodes:

[0040] ;

[0041] wherein, is the heat capacity matrix, determined by and c p ; is the heat conduction matrix, determined by the thermal conductivity; is the heat load vector;

[0042] ​The time derivative term is discretized, and the control equation for time step n+1 becomes a linear equation group:

[0043] ;

[0044] where, is the time step; is the heat load vector at time n+1, including the volume heat source term q and the influence of boundary conditions; is the temperature node vector at time n;

[0045] At each time step, the linear equation group is solved to obtain the temperature of all grid nodes at the corresponding time ; the battery core temperature is the temperature value at the geometric center node of the battery.

[0046] Compared with the prior art, the beneficial effects of the present application are:

[0047] 1. The present application does not need to implant sensors in the battery, directly measures the surface heat flow by multi-point arrangement of heat flow sensors, eliminates the thermal inertia of the traditional compensation method, can respond more quickly and more accurately to the transient change of the battery heating power, ensures the safety of the test and the originality of the battery state, is suitable for the test of commercialized battery cells, can also preliminarily judge the heating uniformity of the battery, provides data support for three-dimensional temperature field reconstruction, and is helpful for positioning the local overheating point.

[0048] 2. The present application combines the directly measured heat flow (reflecting the internal heat generation intensity) and the surface temperature to provide extremely critical and accurate input conditions for the heat model, so that the internal temperature calculated by inversion is much more accurate than the estimated model based on the surface temperature alone. DETAILED DESCRIPTION

[0049] In order to more clearly illustrate the technical solutions of the present application, the following briefly introduces the drawings needed in the embodiments. Obviously, the drawings described in the following embodiments are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0050] Figure 1 is a structural schematic view of a lithium battery internal temperature and heating power online detection system based on heat flow and surface temperature in an embodiment of the present application;

[0051] Figure 2 is a schematic view of a battery clamp and a heat flow sensing module (showing a flexible heat conduction gasket design) in an embodiment of the present application;

[0052] Figure 3A flow chart of a lithium battery internal temperature and heating power online detection method based on heat flow and surface temperature in an embodiment of the present application;

[0053] Figure 4 A battery surface temperature, current and heating power curve graph measured in an embodiment of the present application;

[0054] Figure 5 A battery internal temperature graph obtained by finite element simulation in an embodiment of the present application;

[0055] Wherein, 1—isothermal environment cabin; 2—battery clamp; 3—data acquisition card; 4—main control system; 5—heating / cooling unit; 6—high-precision surface temperature sensor; 7—flexible heat flow sensor layer; 8—lithium battery; 9—electrical load module; 10—flexible heat-conducting gasket; 11—protective and insulating outer layer; 12—heat-conducting and interface filling layer; 13—flexible substrate. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0057] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0058] Embodiment one

[0059] The present application provides a lithium battery internal temperature and heating power online detection system based on heat flow and surface temperature, comprising: an isothermal environment cabin 1, a battery clamp 2, a data acquisition card 3, a main control system 4, a heating / cooling unit 5, a high-precision surface temperature sensor 6, a flexible heat flow sensor layer 7, a lithium battery 8, an electrical load module 9 and a flexible heat-conducting gasket 10.

[0060] The isothermal environment cabin 1 is a double-layer insulation structure airtight test cabin, which is used to provide a stable and uniform constant temperature environment. The inner wall of the cabin body is provided with a heating / cooling unit 5, which is controlled by the main control system 4 and maintains the temperature in the cabin at a set value (T_set) through closed-loop adjustment.

[0061] The battery clamp 2 is arranged inside the isothermal environment cabin 1 and is used to fix the lithium battery 8, so as to ensure the position stability of the battery during the test and maintain good thermal contact.

[0062] The lithium battery 8 is the test object, and a high-precision surface temperature sensor 6 and a flexible heat flow sensor layer 7 are attached to the main surface of the lithium battery 8 respectively. The high-precision surface temperature sensor 6 adopts a thin-film platinum resistance or an ultrathin thermocouple structure and is used for measuring the surface temperature of the battery in real time; the flexible heat flow sensor layer 7 is a thermopile array based on the Seebeck effect, can convert the heat flow density q''_surface vertically through the surface of the battery into an electrical signal output, and realizes high-sensitivity and fast-response heat flow measurement.

[0063] The flexible heat-conducting gasket 10 is arranged between the surface of the battery and the sensor layer and is used for filling micro gaps, reducing contact thermal resistance, improving measurement stability and heat conduction efficiency.

[0064] The electrical load module 9 is connected to the lithium battery 8 through a wire and is used for applying charging and discharging current in the experiment process to simulate the battery heating behavior under different working conditions.

[0065] The data acquisition card 3 is used for synchronously collecting signals from the high-precision surface temperature sensor 6 and the flexible heat flow sensor layer 7 and transmitting the collected data to the host system 4 in real time. The host system 4 performs data processing and calculation, including:

[0066] (1) calculating the instantaneous total heating power of the battery based on the heat flow density signal;

[0067] (2) combining the battery surface temperature, the ambient temperature and the heating power data, and calculating the internal temperature distribution and the core temperature of the battery based on the thermal model inversion. The specific process is as follows:

[0068] Direct calculation of the heating power:

[0069] The instantaneous total heating power of the battery (W) is obtained by integrating the surface heat flow density:

[0070] ;

[0071] Wherein q''_surface (W / m²) is the heat flow density measured by the heat flow sensor; A (m²) is the effective heat dissipation area of the battery in contact with the sensor. For the case that the sensors are arranged on n discrete surfaces, the formula calculation can be simplified as:

[0072] ;

[0073] Wherein is the battery surface heat flow density measured by the i th heat flow sensor, W / m²; is the effective heat dissipation area of the i th heat flow sensor in close contact with the surface of the battery, m²; and n is the number of sensors. W is the instantaneous total heat generation of the battery.

[0074] This method eliminates the thermal inertia of the traditional compensation method by directly measuring the heat flow, and can quickly respond to the transient changes of the battery heat generation.

[0075] Internal temperature inversion:

[0076] To achieve high-precision internal temperature inversion, we first establish a three-dimensional thermal model of the battery to capture the real heat distribution. A one-dimensional model can be used as a simplification in specific scenarios.

[0077] A. Three-dimensional thermal model construction using finite element analysis method:

[0078] 1) When creating a three-dimensional solid geometric model, it needs to be constructed based on the actual internal physical structure of the lithium battery 8 to be measured. The model specifically includes: the battery core as the main heat generation and conduction area, the tab as an important heat conduction path and potential local heat source, and the shell or packaging layer with different thermal physical properties than the battery core material. By accurately reproducing the spatial form, relative position and assembly relationship of the above components, it ensures that the heat conduction path in the model is highly consistent with the physical heat path of the real battery.

[0079] 2) Discretize the battery geometry

[0080] The discretization process is achieved through finite element analysis method, which aims to convert the continuous partial differential control equation into a calculable discrete algebraic system, specifically including the following steps:

[0081] ① Spatial discretization: using the weighted residual method, the transient heat conduction equation is discretized on each grid element. First, use the element shape function to construct the approximate expression of the element temperature field; then, substitute the approximate solution into the control equation and set its weighted integral on the element domain to zero, thereby converting the continuous partial differential equation into an element-level matrix equation, i.e.:

[0082] ;

[0083] Where, is the element heat capacity matrix, is the element heat conduction matrix, is the element heat load vector, is the element temperature vector.

[0084] ② System assembly: according to the node connection relationship, the matrix equations of all grid elements are assembled into the global system equation that describes the entire battery temperature field:

[0085] ;

[0086] Where is the global heat capacity matrix, is the global heat conduction matrix, is the global heat load vector, is the temperature vector of all nodes.

[0087] ③ Time discretization: Discretize the time derivative term in the global system equation. Use implicit integration method to convert the equation into a linear algebraic equation group that needs to be solved at each time step:

[0088] ;

[0089] where, is the time step.

[0090] The equation can be expressed in the form of .

[0091] ④ Real-time solution and inversion: At each time step, the real-time collected surface temperature measurement is taken as the boundary condition, and the instantaneous total heating power is taken as the internal heat source to update the right-hand side term of the linear equation group , and the temperature of all grid nodes at this moment is obtained by solving the equation group . The battery core temperature is determined by the node temperature value located at the geometric center of the battery.

[0092] 3) Assign material thermophysical properties to all grid elements.

[0093] Functions / parameters:

[0094] ① Density ρ (kg / m 3 ).

[0095] ② Constant pressure specific heat cp (J / kg·K).

[0096] ③ Thermal conductivity k (W / m·K). For anisotropic materials, k is a tensor.

[0097] 4) Control equation:

[0098] On each element, follow the law of conservation of energy, which is mathematically expressed as the transient heat conduction equation:

[0099] ;

[0100] where is the density of the battery material, kg / m 3 ; is the specific heat capacity of the battery material, J / (kg·K); T(x, y, z, t) is the temperature, which is a function of spatial coordinates (x, y, z) and time t; t is the time, s; k is the thermal conductivity of the battery material, W / m·K; is the internal heat source generation rate, W / m 3 , that is, the heat generation rate per unit volume, is the gradient operator.

[0101] The transient heat conduction equation is directly imposed on each mesh element by the weighted residual principle of the finite element analysis method, which is specifically manifested as establishing a local governing equation at the element level and forming a global system by assembling. At each mesh element, the transient heat conduction equation is imposed in a weak form (integral form). By the element shape function and the weighted residual method, the partial differential equation is converted into an algebraic relationship. For example, for element (e), the governing equation is rewritten as:

[0102] ;

[0103] where, is the region;

[0104] The diffusion term is handled by the partial integration, the boundary condition is introduced, and finally the element matrix equation is obtained.

[0105] 5) Boundary condition and initial condition setting:

[0106] Boundary condition: On all the outer surfaces of the model, a first type of boundary condition (Dirichlet condition) is applied, that is, the measured surface temperature Tsurface(x, t) at the spatial coordinate position x and time t is directly imposed. This ensures strong coupling between the model and the experimental measurement on the boundary.

[0107] ;

[0108] where is the temperature distribution at the outer boundary of the model, K; is the measured battery surface temperature by the surface temperature sensor, K.

[0109] Initial condition: The temperature of the entire model at the starting time is set, which is usually the set temperature of the isothermal environment chamber 1:

[0110] ;

[0111] where is the temperature field distribution of the model at the initial time t=0, K; is the set temperature of the isothermal environment chamber 1, K.

[0112] Source term loading:

[0113] The calculated instantaneous total heat generation rate P heat is introduced into the model to convert it into the volume heat generation rate Under the assumption of uniform heat generation, there is the equation:

[0114] ;

[0115] where V is the battery volume. This value is loaded as a source term onto all (or specified region) volume elements.

[0116] B. Equation solving and internal temperature calculation process:

[0117] 1) Using the finite element analysis method, the control equation is weighted residual approximation on each grid element, and the continuous partial differential equation is converted into a large ordinary differential equation group about all node temperatures :

[0118] ;

[0119] where is the heat capacity matrix, determined by and c p ; is the heat conduction matrix, determined by the thermal conductivity coefficient; is the heat load vector, which contains the source term and the information of the boundary condition.

[0120] Discretize the time derivative term (such as using the backward Euler method, Crank-Nicolson method, etc.). For time step n+1, the control equation becomes:

[0121] ;

[0122] where is the time step, s; is the heat load vector at time n+1, containing the volume heat source term q and the influence of the boundary condition, W; is the temperature node vector of the previous step (time n), K.

[0123] This is a linear equation group .

[0124] where A is the system coefficient matrix, composed of C / Δt and K; is the equivalent load vector, determined by the temperature field of the previous time step and the current heat source;

[0125] At each time step, solve this linear equation group to get the temperature of all grid nodes at this time. The battery core temperature = T(x=L / 2, t), K. That is, the temperature value of the node located at the geometric center of the battery. By looping all time steps, the complete curve of can be obtained.

[0126] In the present application, the "geometric center of the battery" refers to the position of the geometric center determined in the three-dimensional model according to the actual physical structure of the battery. When establishing the three-dimensional finite element model of the battery, the position corresponds to one or more mesh nodes. Core temperature That is, the temperature value at the node is taken. For a common square lithium ion battery, its geometric center is usually located at the spatial center of the battery body; for a cylindrical battery, it is the intersection of its axial and radial centers.

[0127] After the global temperature vector is solved, the core temperature value at that moment can be directly obtained by querying the index of the corresponding geometric center node in the vector. If the geometric center is located between multiple nodes, the core temperature value can be obtained by interpolating the temperatures of the adjacent nodes.

[0128] The above process is realized in the algorithm by pre-set node number or coordinate matching, ensuring that the core temperature value at each time step can be automatically extracted.

[0129] C. One-dimensional simplified model (taking an infinite plate as an example):

[0130] For a square battery with a thickness of L, it can be simplified as a one-dimensional model along the thickness direction (x direction).

[0131] The control equation is simplified as:

[0132] ;

[0133] Boundary conditions:

[0134] ;

[0135] Wherein and are the surface temperature boundary conditions of the battery at x=0 and x=L, respectively.

[0136] Solution: spatial and temporal discretization can be performed using the finite difference method, and the solution process is similar to the three-dimensional finite element analysis method, but the calculation amount is greatly reduced, which is suitable for occasions with high real-time requirements.

[0137] In this embodiment, the heat flow sensing module is a flexible heat-conducting gasket, which is internally embedded with a heat flow sensor and a temperature sensor.

[0138] Embodiment Two

[0139] As shown in Figure 3 , the present application also provides a lithium battery internal temperature and heating power online detection method based on heat flow and surface temperature, which is realized by the system of embodiment one, and the method comprises:

[0140] Step 1: Place the lithium battery 8 attached with high-precision surface temperature sensor 6 and flexible heat flow sensor layer 7 in the isothermal environment cabin 1;

[0141] Step 2: Control the temperature of the isothermal environment cabin 1 to stabilize to the set value;

[0142] Step 3: Apply an electrical load to the lithium battery 8;

[0143] Step 4: Collect the heat flux density signal and temperature signal of the surface of the lithium battery 8 in real time and synchronously;

[0144] Step 5: Calculate the instantaneous total heat generation power of the battery according to the heat flux density signal;

[0145] Step 6: Combine the temperature signal, ambient temperature, and instantaneous total heat generation power to calculate the internal temperature of the battery based on thermal model inversion.

[0146] In this embodiment, the method for calculating the instantaneous total heat generation power of the battery according to the heat flux density signal is:

[0147] ;

[0148] wherein, is the heat flux density measured by the heat flow sensor; A is the effective heat dissipation area of the battery in contact with the sensor.

[0149] In this embodiment, the method for constructing the thermal model is:

[0150] According to the actual size of the lithium battery 8 to be measured, create a three-dimensional entity geometric model thereof, and use the finite element analysis method to discretize the battery geometry into a plurality of volume units, i.e. mesh units;

[0151] Assign material thermal physical properties to all mesh units;

[0152] On each mesh unit, set the control equation, i.e. the transient heat conduction equation;

[0153] On each mesh unit, set the boundary conditions and initial conditions and source term loading, and finally complete the construction of the thermal model.

[0154] In this embodiment, the transient heat conduction equation is:

[0155] ;

[0156] wherein, is the density of the battery material; is the specific heat capacity of the battery material; T is the temperature; t is time; k is the thermal conductivity of the battery material; is the internal heat source generation rate.

[0157] In the present embodiment, the method for calculating the internal temperature of the battery based on thermal model inversion includes:

[0158] The finite element analysis method is used to perform weighted residual approximation on the control equation in each grid cell, to convert the continuous partial differential equation into a set of ordinary differential equations about the temperature of all nodes:

[0159]

[0160] wherein, is a heat capacity matrix, determined by and c p ; is a heat conduction matrix, determined by the thermal conductivity coefficient; is a heat load vector, containing information of the source term and the boundary condition;

[0161] The time derivative term is discretized, and for the time step n+1, the control equation becomes a linear equation set:

[0162]

[0163] At each time step, the linear equation set is solved to obtain the temperature of all grid nodes at the corresponding time ; the battery core temperature is the temperature value of the node located at the geometric center of the battery.

[0164] The specific implementation process is as follows: Step 1. Install the lithium battery 8 to be measured in the clamp, and ensure that the heat flow sensing module is in good contact with the surface of the battery;

[0165] Step 2. Set the target temperature (T_set) of the isothermal environment chamber 1, start the system, and wait for the environmental temperature to stabilize;

[0166] Step 3. Perform charge-discharge test or run other working conditions on the battery;

[0167] Step 4. The data acquisition system synchronously and in real time acquires the heat flux density signal (q''_surface,i) and the corresponding surface temperature signal (T_surface,i) output by each heat flow sensor;

[0168] Step 5. The heat generation power calculation module integrates and calculates the total instantaneous heat generation power (P_heat) of the battery according to the data of all sensors;

[0169] ​​​Step 6. The internal temperature inversion module takes P_heat, T_surface (multiple), T_set as input, runs the thermal model algorithm, and estimates the internal temperature (T_core) of the battery in real time.

[0170] Step 7. Output and record the heating power curve and internal temperature curve.

[0171] Example Three

[0172] This embodiment provides a lithium battery internal temperature and heating power system based on heat flow and surface temperature as shown in Figure 1 、 Figure 2 Based on a commercial isothermal calorimeter, remove or bypass the heater near the battery clamp in the isothermal environment chamber 1 for compensation calculation, customize a dedicated battery clamp with precise mechanical structure (such as spring-loaded or air pressure device) that can ensure the flexible heat flow-temperature composite sensor to be tightly attached to the two large faces of the battery with constant and uniform pressure, to ensure direct measurement of the heat flow emitted from the battery surface, fast response, no delay; At the same time, the weak analog signal output by the sensor (heat flow sensor is micro-voltage, temperature sensor is resistance or millivolt signal) is connected to a high-speed, high-precision, multi-channel synchronous data acquisition card (NI PXIe system).

[0173] The flexible heat flow-temperature composite sensor adopts a top-down layered stacking structure, which has good flexibility and can be attached to the curved surface of square or cylindrical batteries. Its overall structure consists of the following functional layer sequence:

[0174] 1. Protective and insulating outer layer 11 made of polyimide film or other high-performance polymers.

[0175] Used to protect the internal precise sensing elements from physical scratches or impacts; prevent the sensor from making electrical contact with the external environment or battery shell, ensuring test safety; at the same time, it has the characteristics of moisture resistance and chemical corrosion resistance, ensuring the long-term stability of the sensor under complex working conditions.

[0176] 2. High-precision temperature sensor layer, using thin film technology to make high-precision platinum resistance (such as PT100) or ultra-thin T-type thermocouple. This layer is integrated on top of the heat flow sensor and as close to its measurement area as possible. Used to directly and real-time measure the true temperature (T surface ) of the battery and sensor contact interface, providing boundary conditions for subsequent thermal model inversion calculation.

[0177] 3. Flexible heat flux sensor layer 7, the core is a thermopile structure based on the Seebeck effect, composed of dozens of pairs or even hundreds of pairs of micro thermocouples (usually copper-constantan or other materials) connected in series, and these thermocouple junctions are arranged on the upper and lower planes of a flexible substrate (such as polyimide). Used to sense heat flux density, when heat passes vertically through the sensor, a small temperature difference will be generated in the thickness direction of the sensor ). The thermopile converts this temperature difference into a proportional millivolt-level voltage signal (V). The output voltage V is calibrated after leaving the factory, which can be directly converted into heat flux density value (unit: W / m²), the relationship is ; where K is the calibration coefficient of the sensor.

[0178] 4. Heat conduction and interface filling layer 12 is composed of high thermal conductivity flexible silicone pad, thermal conductive paste or low melting point phase change material. Used to fill the air gap caused by all micro irregularities between the sensor and the battery surface, thereby greatly reducing the contact thermal resistance; Ensure that the heat generated on the surface of the battery can be efficiently and losslessly transmitted to the sensing unit, improve the response speed and accuracy of the measurement; And give the sensor overall flexibility, so that it can conform to the slight deformation of the battery surface, and realize close fitting.

[0179] 5. Flexible substrate 13, composed of polyimide (Polyimide) film. As the structural basis of the entire sensor, it carries all functional layers and ensures that the sensor is electrically insulated from the surface of the battery.

[0180] Set the temperature of the isothermal environment chamber 1 to 25°C. After the temperature is stable, discharge the battery at 1C constant current. The flexible heat flux-temperature composite sensor directly measures the heat flux density emitted through the surface of the battery to the isothermal environment chamber 1. At the same time, the flexible heat flux-temperature composite sensor measures the real temperature of the battery surface, and the high-speed data acquisition card synchronously acquires the voltage signals (converted into heat flux density q'') and temperature signals (T_s1, T_s2) of the two heat flux sensors at a frequency of 1Hz. Transferred to the central processing unit.

[0181] The heat power calculation module and the internal temperature inversion module cooperatively calculate the instantaneous total heat power and the internal core temperature of the battery through a joint algorithm based on real-time parameter identification and adaptive Kalman filtering.

[0182] The specific process of the joint algorithm is as follows:

[0183] 1. Heat power calculation module:

[0184] Input: heat flux density of each sensor and the effective area covered .

[0185] Process: Perform operation: This calculation is performed in real-time at each sampling period (e.g. 1 Hz).

[0186] Output: Instantaneous total heat generation of the battery .

[0187] 2. Internal temperature inversion module - based on real-time parameter identification and adaptive unscented Kalman filter (AUKF):

[0188] This module is the core of the innovation, aiming to dynamically correct model errors and optimally estimate internal states. It consists of two coupled sub-modules:

[0189] A. Real-time model parameter identification sub-module

[0190] Function: Online identification of the battery's equivalent thermal conductivity k to address the model mismatch problem caused by changes in battery state (e.g. temperature, SOC).

[0191] Input: Real-time measured surface temperature , instantaneous total heat generation and heat flux density .

[0192] Process:

[0193] 1) Construct the forward model: Establish a one-dimensional flat plate heat conduction forward model based on the currently assumed thermal conductivity k and heat generation , which can predict the surface temperature , and surface heat flux .

[0194] 2) Define the loss function: Construct a multi-objective loss function that combines temperature and heat flux fitting errors:

[0195] ;

[0196] where is the multi-objective loss function, representing the error between the model prediction and the measured value; is the temperature error weight coefficient; is the heat flux error weight coefficient; and are the measured surface temperatures on both sides, K; is the model calculated heat flux density, W / m²; is the heat flux density measured by the heat flux sensor, W / m².

[0197] 3) Optimization: At each time window (e.g. every 10 seconds), use gradient descent or particle swarm optimization to solve for the optimal thermal conductivity :

[0198] ;

[0199] Output: Real-time updated equivalent thermal conductivity that best reflects the current state of the battery .

[0200] B. Adaptive Unscented Kalman Filter (AUKF) State Estimation Submodule

[0201] Function: Using the accurate model provided by the parameter identification submodule ( ), and in the presence of measurement noise and process noise, optimally estimate the temperature field inside the battery, and finally output the core temperature.

[0202] Input:

[0203] Accurate model from parameter identification submodule .

[0204] Real-time measured surface temperature (as an observation).

[0205] Instantaneous total heating power (as a control input).

[0206] Process:

[0207] 1) State-space modeling:

[0208] Discretize the one-dimensional thermal model of the battery (control equation: ) through finite difference method.

[0209] State equation: ;

[0210] where, is the state vector (temperature of all discrete nodes), is the input vector , is the nonlinear state transition function defined by the discrete heat conduction equation, is the process noise, and V is the total volume of the battery, m³.

[0211] Observation equation: ;

[0212] where, is the observation vector (surface temperature measurement), H is the observation matrix, is the observation noise.

[0213] 2) AUKF iterative estimation:

[0214] Prediction step: Based on the state estimation of the last time and the current model (using ), the predicted value of the state at the current time and its covariance are calculated through the unscented transformation.

[0215] Update step: Obtain the new surface temperature measurement , then calculate the Kalman gain and update the state estimation to obtain the optimal internal temperature field estimation .

[0216] Adaptive mechanism: The algorithm estimates and adjusts the process noise covariance matrix Q and the observation noise covariance matrix R online through the covariance matching technique, so that it can automatically adapt to the noise level changes under different working conditions, ensuring the stability and estimation accuracy of the filter.

[0217] Output: The calculated temperature at the center of the battery thickness , that is, the temperature value of the corresponding center node in the state vector .

[0218] The measured battery surface temperature, current and heat generation power are shown in Figure 4 , and the battery internal temperature obtained by finite element simulation is shown in Figure 5 . The heat generation power curve measured by the method of the present application can capture the transient heat generation peak at the initial stage of discharge better than the traditional compensation power curve; more importantly, the center temperature calculated by real-time parameter identification and AUKF algorithm is highly consistent with the result simulated by high-precision finite element software, even under complex working conditions of battery internal resistance change and heat conduction performance degradation, which proves that the method can still maintain high accuracy in dynamically changing environment, and additionally provides the equivalent thermal conductivity coefficient change curve as a new dimension information reflecting the state of health (SOH) of the battery.

[0219] The above-described embodiments are only descriptions of the preferred modes of the present application and do not limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements to the technical solutions of the present application made by those skilled in the art shall fall within the protection scope determined by the claims of the present application.

Claims

1. An online detection system for the internal temperature and heating power of a lithium battery based on heat flow and surface temperature, characterized in that, The system includes: an isothermal environment chamber, a battery clamp, at least one heat flow sensor module, a data acquisition card, a main control system, a heating / cooling unit, a lithium battery, a high-precision surface temperature sensor, a flexible heat flow sensor layer, an electrical load module, and a flexible thermal conductive pad. The isothermal environment chamber is used to provide a constant temperature environment. Its inner wall is equipped with a heating / cooling unit, which is controlled by the main control system to maintain the set temperature. The battery clamp is used to secure the lithium battery; The lithium battery is the test object, and a high-precision surface temperature sensor and a flexible heat flow sensor layer are respectively attached to its surface. The high-precision surface temperature sensor adopts a thin-film platinum resistance or ultra-thin thermocouple structure for real-time measurement of battery surface temperature. The flexible heat flux sensor layer is a thermopile array based on the Seebeck effect, used to convert the heat flux density passing vertically through the battery surface into an electrical signal output. The electrical load module is connected to the lithium battery via wires and is used to apply charging and discharging current during the experiment to simulate the battery heating behavior under different working conditions. The flexible thermally conductive pad is disposed between the battery surface and the sensor layer to fill the microscopic gaps. The data acquisition card is used to synchronously acquire heat flux density and temperature signals in real time; The main control system is used to calculate the instantaneous total heat generation power of the battery based on the heat flux density signal; and to calculate the internal temperature of the battery based on the thermal model by combining the temperature signal, ambient temperature and instantaneous total heat generation power. Using the instantaneous total heat generation power as the internal heat source, inversion calculations are performed based on the transient heat conduction equation; The transient heat conduction equation is: ； in, It refers to the density of the battery material; t is the specific heat capacity of the battery material; T is the temperature; t is the time; k is the thermal conductivity of the battery material. It is the internal heat source generation rate. This is the gradient operator.

2. A method for online detection of internal temperature and heating power of a lithium battery based on heat flow and surface temperature, wherein the method is implemented by the system described in claim 1, characterized in that, The method includes: Step 1: Place the lithium battery with the attached high-precision surface temperature sensor and flexible heat flow sensor layer in an isothermal environment chamber; Step 2: Control the temperature of the isothermal environment chamber to stabilize to the set value; Step 3: Apply an electrical load to the lithium battery; Step 4: Real-time synchronous acquisition of heat flux density and temperature signals on the surface of the lithium battery; Step 5: Calculate the instantaneous total heat generation power of the battery based on the heat flux density signal; Step 6: Combine the temperature signal, ambient temperature, and instantaneous total heat generation power to calculate the internal temperature of the battery based on the thermal model inversion. Using the instantaneous total heat generation power as the internal heat source, inversion calculations are performed based on the transient heat conduction equation; The transient heat conduction equation is: ； in, It refers to the density of the battery material; t is the specific heat capacity of the battery material; T is the temperature; t is the time; k is the thermal conductivity of the battery material. It is the internal heat source generation rate. This is the gradient operator.

3. The method according to claim 2, characterized in that, The method for calculating the instantaneous total heat generation power of a battery based on the heat flux density signal is as follows: ； in, A is the heat flux density measured by the heat flux sensor; A is the effective heat dissipation area in contact between the battery and the sensor.

4. The method according to claim 3, characterized in that, The method for constructing the thermal model is as follows: Based on the actual dimensions of the lithium battery to be tested, a three-dimensional solid geometric model is created. Using the finite element analysis method, the battery geometry is discretized into several volume elements, i.e., mesh elements. Assign material thermophysical properties to all mesh elements; For each grid cell, the governing equation, namely the transient heat conduction equation, is set; Boundary conditions, initial conditions, and source terms are set for each mesh cell to complete the construction of the thermal model.

5. The method according to claim 2, characterized in that, Methods for calculating the internal temperature of a battery based on thermal model inversion include: By using the finite element method, the governing equations are approximated by weighted residuals on each mesh element, transforming the continuous partial differential equations into a single equation relating to the temperature at all nodes. The system of ordinary differential equations: ； in, The heat capacity matrix is ​​formed by... and c p Decide; It is the heat conduction matrix, determined by the thermal conductivity. It is the thermal load vector; Discretizing the time derivative term, for time step n+1, the governing equations become a system of linear equations: ； in, It is the time step; It is the heat load vector at time n+1, which includes the volumetric heat source term q and the influence of boundary conditions; It is the temperature node vector at time n; At each time step, the system of linear equations is solved to obtain the temperature at all grid nodes at the corresponding time. Battery core temperature This refers to the temperature value located at the geometric center node of the battery.

Images