Charging curve generation method, device and equipment

Abstract

The charging curve generation method comprises the steps that a charging plan corresponding to a target battery is acquired, and the charging plan is used for reflecting current and temperature of the target battery at different moments in the charging process; determining an electrochemical parameter set according to the charging plan and an equation conversion model corresponding to the electrochemical mechanism model; the equation conversion model is obtained by converting an electrochemical equation in the electrochemical mechanism model, and the electrochemical parameter set comprises electrochemical parameters of the target battery at different moments in a plurality of charging processes; according to the electrochemical parameter set and a preset voltage model, a voltage set is determined, the preset voltage model is used for reflecting a mapping relation between the electrochemical parameters and voltages, and the voltage set comprises voltages of the target battery at different moments in the multiple charging processes; and generating a full-life charging curve of the target battery according to the voltage set.

Description

Technical Field

[0001] The present application relates to the field of battery management technology, and in particular to a charging curve generation method, device and equipment. Background Art

[0002] With the severe energy shortage and environmental protection problems, higher requirements are placed on batteries. The new generation of lithium-ion batteries are widely used in various fields, such as new energy vehicles and electronic products, due to their characteristics of less pollution and energy diversification.

[0003] Lithium-ion batteries will age with use. When the aging degree of lithium-ion batteries is serious, it may cause safety hazards. Effective lithium-ion battery evaluation technology helps the battery management system to perceive the battery status in real time, thereby determining the aging degree of lithium-ion batteries and eliminating safety hazards in time. Early detection of problems helps to intervene and manage lithium-ion batteries early. In the early use stage of lithium-ion batteries, predicting the long-term battery status of lithium-ion batteries in the future has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present application provides a charging curve generation method, device and equipment, aiming to predict the future long-term battery state of a lithium-ion battery.

[0005] In a first aspect, a charging curve generating method is provided, comprising:

[0006] Acquire a charging plan corresponding to a target battery, wherein the charging plan is used to reflect the current and temperature of the target battery at different times during the charging process;

[0007] Determine an electrochemical parameter set according to the charging plan and an equation conversion model corresponding to the electrochemical mechanism model, wherein the equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model, wherein the equation conversion model uses the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter and uses the dependent variable in the electrochemical equation as an output parameter, wherein the equation conversion model is used to reflect the mapping relationship between the input parameter and the output parameter, and wherein the electrochemical parameter set includes the electrochemical parameters of the target battery at different times during multiple charging processes;

[0008] Determine a voltage set according to the electrochemical parameter set and a preset voltage model, wherein the preset voltage model is used to reflect a mapping relationship between electrochemical parameters and voltages, and the voltage set includes voltages of the target battery at different times during the multiple charging processes;

[0009] A full-life charging curve of the target battery is generated according to the voltage set.

[0010] In this technical solution, after obtaining a charging plan that reflects the current and temperature of the battery at different times during the charging process, an electrochemical parameter set is determined according to the charging plan and an equation conversion model corresponding to the electrochemical mechanism model, and then a voltage set is determined according to the electrochemical parameter set and a preset voltage model. Finally, a full life curve of the target battery is generated according to the voltage set; since the voltage set includes the voltage of the target battery at different times during multiple charging processes, it is equivalent to predicting the long-term voltage of the battery in the future, and obtaining a charging curve that reflects the long-term battery state in the future is helpful to detect abnormal battery conditions early; since the equation conversion model is composed of The electrochemical equation in the electrochemical mechanism model is converted. The equation conversion model takes the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter, and takes the dependent variable in the electrochemical equation as an output parameter. The equation conversion model is used to reflect the mapping relationship between the input parameters and the output parameters. It is equivalent to converting the complex electrochemical mechanism model into a simple model that reflects the mapping relationship between the independent variable and the dependent variable in the electrochemical equation to realize the solution of the electrochemical parameters at different times in multiple charging processes, thereby improving the solution speed of the electrochemical parameters at different times in multiple charging processes, thereby improving the speed of generating the full life charging curve of the battery.

[0011] In combination with the first aspect, the electrochemical parameter set includes a current density set, a liquid lithium ion concentration set, a liquid potential set, a solid potential set and a solid lithium ion concentration set, the current density set includes the current density of the target battery at different times in the multiple charging processes, the liquid lithium ion concentration set includes the liquid lithium ion concentration of the target battery at different times in the multiple charging processes, the liquid potential set includes the liquid potential of the target battery at different times in the multiple charging processes, the solid potential set includes the solid potential of the target battery at different times in the multiple charging processes, and the solid lithium ion concentration set includes the solid lithium ion concentration of the target battery at different times in the multiple charging processes; the electrochemical parameter set is determined according to the equation conversion model corresponding to the charging plan and the electrochemical mechanism model, including: determining the current density according to the charging plan and the Bavo equation conversion model, the Bavo equation conversion model uses the liquid lithium ion concentration, current and temperature in the Bavo equation as input parameters, and uses the Bavo equation as the input parameters. The current density in the process is used as the output parameter; the liquid phase lithium ion concentration is determined according to the current density and the liquid phase mass equation conversion model, the liquid phase mass equation conversion model uses the current density and time in the liquid phase mass equation as input parameters, and the liquid phase lithium ion concentration in the liquid phase mass equation is used as the output parameter; the liquid phase potential is determined according to the current density, the liquid phase lithium ion concentration and the liquid phase charge equation conversion model, the liquid phase charge equation conversion model uses the liquid phase lithium ion concentration and the current density in the liquid phase charge equation as input parameters, and the liquid phase potential in the liquid phase charge equation is used as the output parameter; the solid phase potential is determined according to the current density and the solid phase charge equation conversion model, the solid phase charge equation conversion model uses the current density in the solid phase charge equation as an input parameter, and the solid phase potential in the solid phase charge equation is used as an output parameter; the solid phase lithium ion concentration is determined according to the particle radius of the target battery and the solid phase mass equation conversion model, the solid phase mass equation conversion model uses the particle radius and time in the solid phase mass equation as input parameters, and the solid phase lithium ion concentration in the solid phase mass equation is used as the output parameter.

[0012] The five equations in the electrochemical mechanism model are converted into corresponding models that reflect the mapping relationship between independent variables and dependent variables, and the converted model is used to solve the current density, liquid lithium ion concentration, liquid potential, solid potential and solid lithium ion concentration in turn to achieve a simple solution to the electrochemical parameters.

[0013] In combination with the first aspect, in a possible implementation method, determining the current density according to the charging plan and the Bavo equation conversion model includes: obtaining the current at the target time and the temperature at the target time from the charging plan, where the target time is any time during the charging process; inputting the current at the target time and the temperature at the target time and the liquid lithium ion concentration at the target time into the Bavo equation conversion model to obtain the current density at the target time.

[0014] The current density at a certain moment is determined by a Bavo equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the Bavo equation. The iteration time and the charging process determine the current density at different moments in multiple processes.

[0015] In combination with the first aspect, in a possible implementation method, determining the liquid phase lithium ion concentration based on the current density and the liquid phase mass equation conversion model includes: inputting the current density at the target moment and the moment next to the target moment into the liquid phase mass equation conversion model to obtain the liquid phase lithium ion concentration at the moment next to the target moment, wherein the target moment is any moment in the charging process.

[0016] The liquid phase lithium ion concentration at the next moment is determined by a liquid phase mass equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the liquid phase mass equation. The iteration time and the charging process determine the liquid phase lithium ion concentration at different moments in multiple processes.

[0017] In combination with the first aspect, in a possible implementation method, determining the liquid phase potential based on the current density, the liquid phase lithium ion concentration and the liquid phase charge equation conversion model includes: inputting the current density at the target moment and the liquid phase lithium ion concentration at the target moment into the liquid phase charge equation conversion model to obtain the liquid phase potential at the target moment, where the target moment is any moment in the charging process.

[0018] The liquid phase potential at a certain moment is determined by a liquid phase charge equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the liquid phase charge equation. The iteration time and charging process determine the liquid phase potential at different moments in multiple processes.

[0019] In combination with the first aspect, in a possible implementation method, determining the solid-phase potential based on the current density and the solid-phase charge equation conversion model includes: inputting the current density at the target moment into the solid-phase charge equation conversion model to obtain the solid-phase potential at the target moment, where the target moment is any moment in the charging process.

[0020] The solid-phase potential at a certain moment is determined by a solid-phase charge equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the solid-phase charge equation. The iteration time and charging process determine the solid-phase potential at different moments in multiple processes.

[0021] In combination with the first aspect, in a possible implementation method, determining the solid-phase lithium ion concentration based on the particle radius of the target battery and the solid-phase mass equation conversion model includes: inputting the target time and the particle radius of the target battery into the solid-phase mass equation conversion model to obtain the solid-phase lithium ion concentration at the target time, where the target time is any time during the charging process.

[0022] The solid phase lithium ion concentration at a certain moment is determined by a solid phase mass equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the solid phase mass equation. The iteration time and charging process determine the solid phase lithium ion concentration at different moments in multiple processes.

[0023] In combination with the first aspect, in a possible implementation method, determining the voltage set based on the electrochemical parameter set and the preset voltage model includes: taking the first charging process as the target charging process, and taking the first moment in the target charging process as the target moment; inputting the electrochemical parameters at the target moment into the preset voltage model to obtain the voltage at the target moment; taking the next moment of the target moment in the target charging process as the target moment, and executing the step of inputting the electrochemical parameters at the target moment into the preset voltage model to obtain the voltage at the target moment, until the target moment is the last moment in the target charging process; taking the next charging process of the target charging process as the target charging process, and executing the step of taking the first moment in the target charging process as the target moment.

[0024] By inputting the electrochemical parameters into a preset voltage model that reflects the mapping relationship between the electrochemical parameters and the voltage, the voltage at a certain moment is obtained, and the time and charging process are continuously iterated to determine the voltage at different moments in multiple charging processes.

[0025] In a second aspect, a charging curve generating device is provided, comprising:

[0026] An acquisition module, used for acquiring a charging plan corresponding to a target battery, wherein the charging plan is used for reflecting the current and temperature of the target battery at different times during the charging process;

[0027] a parameter determination module, for determining a set of electrochemical parameters according to the charging plan and an equation conversion model corresponding to the electrochemical mechanism model;

[0028] The equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model, and the equation conversion model uses the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter, and uses the dependent variable in the electrochemical equation as an output parameter;

[0029] The equation conversion model is used to reflect the mapping relationship between input parameters and output parameters, and the electrochemical parameter set includes electrochemical parameters of the target battery at different times during multiple charging processes;

[0030] a voltage determination module, configured to determine a voltage set according to the electrochemical parameter set and a preset voltage model, wherein the preset voltage model is configured to reflect a mapping relationship between electrochemical parameters and voltages, and the voltage set includes voltages of the target battery at different times during the multiple charging processes;

[0031] A curve generating module is used to generate a full-life charging curve of the target battery according to the voltage set.

[0032] In a third aspect, a computer device is provided, comprising a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory, and when the processor executes the one or more computer programs, the computer device implements the charging curve generation method of the first aspect.

[0033] In a fourth aspect, a battery device is provided, comprising the aforementioned charging curve generating apparatus, or comprising the aforementioned computer device.

[0034] The present application achieves the following technical effects: since the voltage set includes the voltage of the target battery at different times during multiple charging processes, it is equivalent to predicting the long-term voltage of the battery in the future, and obtaining a charging curve that reflects the battery's long-term battery status in the future, which is helpful to detect the abnormal state of the battery early; since the equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model, the equation conversion model uses the independent variables in the electrochemical equation corresponding to the electrochemical mechanism model as input parameters, and uses the dependent variables in the electrochemical equation as output parameters. The equation conversion model is used to reflect the mapping relationship between the input parameters and the output parameters, which is equivalent to converting a complex electrochemical mechanism model into a simple model that reflects the mapping relationship between the independent variables and the dependent variables in the electrochemical equation to realize the solution of the electrochemical parameters at different times in multiple charging processes, thereby improving the solution speed of the electrochemical parameters at different times in multiple charging processes, thereby improving the generation speed of the battery's full life charging curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0036] Figure 1 A schematic diagram of a flow chart of a charging curve generation method provided in an embodiment of the present application;

[0037] Figure 2A-2E Schematic diagrams of the Bavo equation conversion model, the liquid phase mass equation conversion model, the liquid phase charge equation conversion model, the solid phase charge equation conversion model, and the solid phase mass equation conversion model provided in the embodiments of the present application, respectively;

[0038] Figure 3 A schematic diagram of a process for determining a set of electrochemical parameters provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of a preset voltage model provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of a process for determining a voltage set provided in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of a full-life charging curve provided in an embodiment of the present application;

[0042] Figure 7 The process steps of training and obtaining the equation conversion model corresponding to the electrochemical mechanism model provided in the embodiment of the present application;

[0043] Figure 8 It is a structural schematic diagram of a charging curve generating device provided in an embodiment of the present application;

[0044] Fig. 9 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] It is understandable that, if there is no conflict, the various features in the embodiments of the present application can be combined with each other, all within the scope of protection of the present application. In addition, although the functional module division is performed in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a sequence different from the module division in the device or the flow chart. Furthermore, the words "first", "second", "third", etc. used in this application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0047] The technical solution of the present application is applicable to the application scenarios of lithium-ion batteries. The technical solution of the present application is used to predict the full-life charging curve of lithium-ion batteries in the application scenarios of lithium-ion batteries. In the present application, the full-life charging curve of a lithium-ion battery refers to the charging curve of a lithium-ion battery from the time when the lithium-ion battery is put into use to the time when the lithium-ion battery ages and can no longer be used. The technical solution of the present application is applied to equipment containing lithium-ion batteries, such as electric two-wheeled vehicles, electric vehicles, electronic equipment, energy storage equipment, drones, power tools, etc., which are not limited by the present application.

[0048] In order to realize the prediction of the full life charging curve of lithium-ion batteries, the present application constructs an electrochemical mechanism model, and obtains the charging data of lithium-ion batteries under parameters such as different temperatures, different charge and discharge rates, and different ambient temperatures through the electrochemical mechanism model simulation. The charging data includes voltage, current, temperature, current density, solid phase potential, solid phase lithium ion concentration, particle radius, time, liquid phase lithium ion concentration, and liquid phase potential. The electrochemical mechanism model is simplified into an equation conversion model with the independent variables in the electrochemical equation corresponding to the electrochemical mechanism model as input parameters and the dependent variables in the electrochemical equation as output parameters. According to the charging data obtained by simulation, the mapping relationship between the independent variables and the dependent variables represented by the equation conversion model is determined, and the charging data obtained by simulation is used to calculate the charging curve of lithium-ion batteries. A mapping relationship between chemical parameters and real voltage is used to determine a preset voltage model; the electrochemical parameters of the lithium-ion battery at different times in a variety of charging processes are determined according to a predetermined equation conversion model and a charging plan for the lithium-ion battery; then, the voltage of the target battery at different times in a variety of charging processes is determined according to the electrochemical parameters of the lithium-ion battery at different times in a variety of charging processes and a preset voltage model; finally, a full-life charging curve of the lithium-ion battery is generated according to the voltage of the target battery at different times in a variety of charging processes. The full-life curve of the lithium-ion battery reflects the long-term battery status of the lithium-ion battery in the future, helps to predict the aging of the battery in the future, and facilitates timely management of the lithium-ion battery, thereby slowing down the battery aging rate and extending the battery life.

[0049] The technical solution of this application is described in detail below.

[0050] See also Figure 1 , Figure 1 A schematic diagram of a charging curve generation method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method comprises the following steps:

[0051] S101, obtaining a charging plan corresponding to a target battery.

[0052] Here, the target battery is a lithium-ion battery whose charging curve needs to be predicted.

[0053] The charging plan corresponding to the target battery is used to reflect the current and temperature of the target battery at different times during the charging process. In some embodiments, the charging plan corresponding to the target battery is pre-set and saved in the battery management system corresponding to the target battery, and the above charging plan is obtained from the battery management system corresponding to the target battery. For example, the charging plan is a SOP charging function pre-set in the battery management system. It is understandable that the SOP charging function is presented in a table, curve or function form.

[0054] S102, determining a set of electrochemical parameters according to a charging plan corresponding to the target battery and an equation conversion model corresponding to the electrochemical mechanism model.

[0055] Here, the equation conversion model corresponding to the electrochemical mechanism model is obtained by converting the electrochemical equation in the electrochemical mechanism model. The equation conversion model corresponding to the electrochemical mechanism model takes the independent variables in the electrochemical equation corresponding to the electrochemical mechanism model as input parameters, and takes the dependent variables in the electrochemical equation as output parameters. The equation conversion model is used to reflect the mapping relationship between the input parameters and the output parameters.

[0056] In some embodiments of the present application, the electrochemical mechanism model is described as 5 equations, which are the Bavo equation, the liquid mass equation, the solid charge equation, the liquid charge equation, and the solid mass equation. It is understandable that the Bavo equation, the liquid mass equation, the solid charge equation, the liquid charge equation, and the solid mass equation have multiple forms of expression.

[0057] Exemplarily, the Bavo equation is expressed as:

[0058]

[0059] j represents current density, I represents current, R represents molar gas constant, T represents temperature, α a and α c represents the transfer coefficient, F represents the Faraday constant, η represents the surface overpotential, c e represents the concentration of lithium ions in the liquid phase, c s，max represents the maximum concentration of solid phase lithium ions, c s，eRepresents the lithium ion concentration at the solid-liquid interface.

[0060] Exemplarily, the liquid phase mass equation is expressed as:

[0061]

[0062] ε e represents the liquid volume fraction, t represents time, c e Denotes the concentration of lithium ions in the liquid phase, De eff represents the effective diffusion coefficient of the liquid phase, a s It represents the active surface area per unit volume of the electrode.

[0063] Exemplarily, the liquid phase charge equation is expressed as:

[0064]

[0065] k eff Indicates the effective conductivity of the liquid phase, Kd eff is the liquid phase effective diffusion coefficient, represents the liquid phase potential.

[0066] Exemplarily, the solid phase charge equation is expressed as:

[0067]

[0068] σ eff is the solid phase effective conductivity, Represents the solid-phase electromotive force.

[0069] Exemplarily, the solid phase mass equation is expressed as:

[0070]

[0071] c s represents the solid phase lithium ion concentration, Ds represents the solid phase diffusion coefficient, and r represents the particle radius.

[0072] The equation conversion models corresponding to the electrochemical mechanism model include the Bavo equation conversion model, the liquid phase mass equation conversion model, the liquid phase charge equation conversion model, the solid phase charge equation conversion model and the solid phase mass equation conversion model.

[0073] The Bavo equation conversion model uses the liquid phase lithium ion concentration, current and temperature in the Bavo equation as input parameters, and the current density in the Bavo equation as output parameter. The Bavo equation conversion model is as follows: Figure 2A The Bavo equation conversion model is obtained by pre-training. The specific process of training the Bavo equation conversion model will be described in detail later.

[0074] In some embodiments, the Bavo equation conversion model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0075] Taking the Bavo equation conversion model as MLP (multi-layer perception) as an example, the Bavo equation conversion model is as follows Figure 2A As shown in m1 in the figure, the Bavo equation conversion model includes an input layer, a hidden layer and an output layer. The input layer of the Bavo equation conversion model is used to input the liquid phase lithium ion concentration c e , current I and temperature T, the hidden layer of the Bavo equation conversion model is used to convert the liquid phase lithium ion concentration c into the input layer e , current I and temperature T are processed to obtain hidden features. The connection matrix between the hidden layer of the Bavo equation conversion model and the input layer of the Bavo equation conversion model is 3*Q1, Q1 is the number of neurons in the hidden layer of the Bavo equation conversion model, that is, the data dimension of the hidden features output by the hidden layer of the Bavo equation conversion model; the output layer of the Bavo equation conversion model is used to output the current density j according to the hidden features output by the hidden layer of the Bavo equation conversion model, and the connection matrix between the output layer of the Bavo equation conversion model and the hidden layer of the Bavo equation conversion model is Q1*1. It can be understood that Figure 2A The m1 in the figure is only an example given in this application to explain the conversion model of the Bavo equation and does not limit this application.

[0076] In some embodiments, the liquid mass equation conversion model uses the current density and time in the liquid mass equation as input parameters and the liquid lithium ion concentration in the liquid mass equation as output parameter. The liquid mass equation conversion model is as follows: Figure 2B The liquid phase mass equation conversion model is obtained by pre-training, and the specific process of training the liquid phase mass equation conversion model will be described in detail later.

[0077] In some embodiments, the liquid mass equation conversion model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0078] Taking the liquid phase mass equation conversion model as MLP as an example, the liquid phase mass equation conversion model is as follows Figure 2BAs shown in m2 in the figure, the liquid mass equation conversion model includes an input layer, a hidden layer and an output layer. The input layer of the liquid mass equation conversion model is used to input current density j and time t. The hidden layer of the liquid mass equation conversion model is used to process the current density j and time t input by the input layer to obtain hidden features. The connection matrix between the hidden layer of the liquid mass equation conversion model and the input layer of the liquid mass equation conversion model is 2*Q2. Q2 is the number of neurons in the hidden layer of the liquid mass equation conversion model, that is, the data dimension of the hidden features output by the hidden layer of the liquid mass equation conversion model. The output layer of the liquid mass equation conversion model is used to output the liquid lithium ion concentration c according to the hidden features output by the hidden layer of the liquid mass equation conversion model. e , the connection matrix between the output layer of the liquid mass equation conversion model and the hidden layer of the liquid mass equation conversion model is Q2*1. It can be understood that, Figure 2B The m2 in this example is only an example given in this application to explain the liquid phase mass equation conversion model and does not limit this application.

[0079] In some embodiments, the liquid phase charge equation conversion model uses the liquid phase lithium ion concentration and current density in the liquid phase charge equation as input parameters, and uses the liquid phase potential in the liquid phase charge equation as output parameter. The liquid phase charge equation conversion model is as follows: Figure 2C The liquid phase charge equation conversion model is obtained by pre-training, and the specific process of training the liquid phase charge equation conversion model will be described in detail later.

[0080] In some embodiments, the liquid phase charge equation conversion model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0081] Taking the liquid phase charge equation conversion model as MLP as an example, the liquid phase charge equation conversion model is as follows Figure 2C As shown in m3 in FIG. 1 , the liquid phase charge equation conversion model includes an input layer, a hidden layer, and an output layer. The input layer of the liquid phase charge equation conversion model is used to input the liquid phase lithium ion concentration c e and current density j, the hidden layer of the liquid charge equation conversion model is used to convert the liquid lithium ion concentration c into the input layer e The hidden features are obtained by data processing of the current density j. The connection matrix between the hidden layer of the liquid charge equation conversion model and the input layer of the liquid charge equation conversion model is 2*Q3. Q3 is the number of neurons in the hidden layer of the liquid charge equation conversion model, that is, the data dimension of the hidden features output by the hidden layer of the liquid charge equation conversion model. The output layer of the liquid charge equation conversion model is used to output the liquid potential according to the hidden features output by the hidden layer of the liquid charge equation conversion model. The connection matrix between the output layer of the liquid phase charge equation conversion model and the hidden layer of the liquid phase charge equation conversion model is Q3*1. It can be understood that Figure 2C The m3 in the figure is merely an example given in this application to explain the liquid phase charge equation conversion model and does not limit this application.

[0082] In some embodiments, the solid-phase charge equation conversion model uses the current density in the solid-phase charge equation as an input parameter and the solid-phase potential in the solid-phase charge equation as an output parameter. Figure 2D The solid-phase charge equation conversion model is obtained by pre-training, and the specific process of training the solid-phase charge equation conversion model will be described in detail later.

[0083] In some embodiments, the solid-phase charge equation conversion model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0084] Taking the solid phase charge equation conversion model as MLP as an example, the solid phase charge equation conversion model is as follows Figure 2D As shown in m4 in the figure, the solid-phase charge equation conversion model includes an input layer, a hidden layer and an output layer. The input layer of the solid-phase charge equation conversion model is used to input current density j, and the hidden layer of the solid-phase charge equation conversion model is used to perform data processing on current density j to obtain hidden features. The connection matrix between the hidden layer of the solid-phase charge equation conversion model and the input layer of the solid-phase charge equation conversion model is 1*Q4, and Q4 is the number of neurons in the hidden layer of the solid-phase charge equation conversion model, that is, the data dimension of the hidden features output by the hidden layer of the solid-phase charge equation conversion model; the output layer of the solid-phase charge equation conversion model is used to output the solid-phase potential according to the hidden features output by the hidden layer of the solid-phase charge equation conversion model. The connection matrix between the output layer of the solid-phase charge equation conversion model and the hidden layer of the solid-phase charge equation conversion model is Q4*1. It can be understood that Figure 2D The m4 in this example is only an example given in this application to explain the solid-phase charge equation conversion model and does not limit this application.

[0085] In some embodiments, the solid phase mass equation conversion model uses the particle radius and time in the solid phase mass equation as input parameters, and uses the solid phase lithium ion concentration in the solid phase mass equation as output parameter. The solid phase mass equation conversion model is as follows: Figure 2E The solid phase mass equation conversion model is obtained by pre-training, and the specific process of training the solid phase mass equation conversion model will be described in detail later.

[0086] In some embodiments, the solid-phase mass equation conversion model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0087] Taking the solid phase mass equation conversion model as MLP as an example, the solid phase mass equation conversion model is as follows Figure 2E As shown in m5 in the figure, the solid-phase mass equation conversion model includes an input layer, a hidden layer and an output layer. The input layer of the solid-phase mass equation conversion model is used to input the particle radius r and time t. The hidden layer of the solid-phase mass equation conversion model is used to process the particle radius r and time t input by the input layer to obtain hidden features. The connection matrix between the hidden layer of the solid-phase mass equation conversion model and the input layer of the solid-phase mass equation conversion model is 2*Q5. Q5 is the number of neurons in the hidden layer of the solid-phase mass equation conversion model, that is, the data dimension of the hidden features output by the hidden layer of the solid-phase mass equation conversion model. The output layer of the solid-phase mass equation conversion model is used to output the solid-phase lithium ion concentration c according to the hidden features output by the hidden layer of the solid-phase mass equation conversion model. s , the connection matrix between the output layer of the solid mass equation conversion model and the hidden layer of the solid mass equation conversion model is Q5*1. It can be understood that, Figure 2E The m5 in the figure is only an example given in this application to explain the solid phase mass equation conversion model and does not limit this application.

[0088] The electrochemical parameter set includes a current density set, a liquid lithium ion concentration set, a liquid potential set, a solid potential set, and a solid lithium ion concentration set. The current density set includes the current density of the target battery at different times during multiple charging processes, the liquid lithium ion concentration set includes the liquid lithium ion concentration of the target battery at different times during multiple charging processes, the liquid potential set includes the liquid potential of the target battery at different times during multiple charging processes, the solid potential set includes the solid potential of the target battery at different times during multiple charging processes, and the solid lithium ion concentration set includes the solid lithium ion concentration of the target battery at different times during multiple charging processes. Determine the electrochemical parameters through the following steps A1-A5:

[0089] A1. Determine the current density based on the charging plan corresponding to the target battery and the Bavo equation conversion model.

[0090] The current at the target time and the temperature at the target time are obtained from the charging plan corresponding to the target battery, and the current at the target time, the temperature at the target time and the liquid lithium ion concentration at the target time are input into the Bavo equation conversion model to obtain the current density at the target time. The target time is any time during the charging process.

[0091] The conversion model based on the Bavo equation is Figure 2A Taking m1 in as an example, the current at the target moment, the temperature at the target moment and the liquid-phase lithium ion concentration at the target moment are input into the Bavo equation conversion model. The current at the target moment, the temperature at the target moment and the liquid-phase lithium ion concentration at the target moment constitute a 1*3 input matrix. The hidden layer in the Bavo equation conversion model multiplies the 1*3 input matrix with the 3*Q1 connection matrix to obtain a 1*Q1 hidden matrix. The output layer in the Bavo equation conversion model multiplies the 1*Q1 hidden matrix with the Q1*1 connection matrix to output the current density at the target moment.

[0092] The current density at a certain moment is determined by a Bavo equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the Bavo equation. The iteration time and the charging process determine the current density at different moments in multiple processes.

[0093] A2. Determine the liquid phase lithium ion concentration based on the current density and liquid phase mass equation conversion model.

[0094] The current density at the target moment and the moment next to the target moment are input into the liquid phase mass equation conversion model to obtain the liquid phase lithium ion concentration at the moment next to the target moment.

[0095] The liquid phase mass equation conversion model is Figure 2B Taking m2 in as an example, the current density at the target moment and the next moment after the target moment are input into the liquid mass equation conversion model. The current density at the target moment and the next moment after the target moment constitute a 1*2 input matrix. The hidden layer in the liquid mass equation conversion model multiplies the 1*2 input matrix with the 2*Q2 connection matrix to obtain a 1*Q2 hidden matrix. The output layer in the liquid mass equation conversion model multiplies the 1*Q2 hidden matrix with the Q2*1 connection matrix to output the liquid lithium ion concentration at the next moment after the target moment.

[0096] The liquid phase lithium ion concentration at the next moment is determined by a liquid phase mass equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the liquid phase mass equation. The iteration time and the charging process determine the liquid phase lithium ion concentration at different moments in multiple processes.

[0097] A3. Determine the liquid phase potential based on the current density, liquid phase lithium ion concentration and liquid phase charge equation conversion model.

[0098] The current density at the target time and the liquid phase lithium ion concentration at the target time are input into the liquid phase charge equation conversion model to obtain the liquid phase potential at the target time.

[0099] The liquid phase charge equation conversion model is Figure 2CTaking m3 in as an example, the current density at the target moment and the liquid-phase lithium ion concentration at the target moment are input into the liquid-phase mass equation conversion model. The current density at the target moment and the liquid-phase lithium ion concentration at the target moment constitute a 1*2 input matrix. The hidden layer in the liquid-phase mass equation conversion model multiplies the 1*2 input matrix with the 2*Q3 connection matrix to obtain a 1*Q3 hidden matrix. The output layer in the liquid-phase charge equation conversion model multiplies the 1*Q3 hidden matrix with the Q3*1 connection matrix to output the liquid-phase potential at the target moment.

[0100] The liquid phase potential at a certain moment is determined by a liquid phase charge equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the liquid phase charge equation. The iteration time and charging process determine the liquid phase potential at different moments in multiple processes.

[0101] A4. Determine the solid phase potential based on the current density and solid phase charge equation conversion model.

[0102] The current density at the target time is input into the solid-phase charge equation conversion model to obtain the solid-phase potential at the target time.

[0103] The solid phase charge equation conversion model is Figure 2D Taking m4 in as an example, the current density at the target time is input into the solid-phase charge equation conversion model. The current density at the target time is a 1*1 input matrix. The hidden layer in the solid-phase charge equation conversion model multiplies the 1*1 input matrix with the 1*Q4 connection matrix to obtain a 1*Q4 hidden matrix. The output layer in the solid-phase charge equation conversion model multiplies the 1*Q4 hidden matrix with the Q4*1 connection matrix to output the solid-phase potential at the target time.

[0104] The solid-phase potential at a certain moment is determined by a solid-phase charge equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the solid-phase charge equation. The solid-phase potential at different moments in multiple processes can be determined by iterating the time and charging process.

[0105] A5. Determine the solid phase lithium ion concentration based on the particle radius of the target battery and the solid phase mass equation conversion model.

[0106] The particle radius of the target time and the target battery are input into the solid mass equation conversion model to obtain the solid lithium ion concentration at the target time.

[0107] The solid phase mass equation conversion model is Figure 2ETaking m5 in as an example, the target time and the particle radius of the target battery are input into the solid-phase mass equation conversion model. The target time and the particle radius of the target battery constitute a 1*2 input matrix. The hidden layer in the solid-phase mass equation conversion model multiplies the 1*2 input matrix with the 2*Q5 connection matrix to obtain a 1*Q5 hidden matrix. The output layer in the solid-phase mass equation conversion model multiplies the 1*Q5 hidden matrix with the Q5*1 connection matrix to output the solid-phase lithium ion concentration at the target time.

[0108] The solid phase lithium ion concentration at a certain moment is determined by a solid phase mass equation conversion model that reflects the mapping relationship between the independent variable and the dependent variable in the solid phase mass equation. The iteration time and charging process determine the solid phase lithium ion concentration at different moments in multiple processes.

[0109] The following describes the process of iteratively obtaining the electrochemical parameter set, see Figure 3 , Figure 3 A schematic diagram of a process for determining a set of electrochemical parameters includes the following steps B1-B13:

[0110] B1. Take the first charging process as the target charging process.

[0111] B2. The first moment in the target charging process is taken as the target moment.

[0112] B3. Taking the initial liquid phase lithium ion concentration as the liquid phase lithium ion concentration at the target time.

[0113] Here, the initial liquid-phase lithium ion concentration is the liquid-phase lithium ion concentration of the target battery, which is obtained by measuring the initial liquid-phase lithium ion concentration of the target battery.

[0114] B4. Obtain the current and temperature at the target time from the charging plan corresponding to the target battery, input the current, temperature and liquid lithium ion concentration at the target time into the Bavo equation conversion model to obtain the current density at the target time.

[0115] B5. Input the current density at the target moment and the moment next to the target moment into the liquid phase mass equation conversion model to obtain the liquid phase lithium ion concentration at the moment next to the target moment.

[0116] B6. Input the current density at the target time and the liquid phase lithium ion concentration at the target time into the liquid phase charge equation conversion model to obtain the liquid phase potential at the target time.

[0117] B7. Input the current density at the target time into the solid-phase charge equation conversion model to obtain the solid-phase potential at the target time.

[0118] B8. Input the particle radius of the target time and the target battery into the solid mass equation conversion model to obtain the solid lithium ion concentration at the target time.

[0119] B9. Determine whether the target time is the last time in the target charging process.

[0120] If the target time is not the last time in the target charging process, execute step B10; if the target time is the last time in the target charging process, execute step B11.

[0121] B10. Take the next moment after the target moment in the target charging process as the target moment, and execute steps B4 and B8.

[0122] B11. The next charging process of the target charging process is used as the target charging process.

[0123] B12. Take the first moment in the target charging process as the target moment, and execute steps B8 and B13.

[0124] B13. Taking the liquid phase lithium ion concentration at the last moment of the previous charging process of the target charging process as the liquid phase lithium ion concentration at the target moment, and executing step B4.

[0125] S103, determining a voltage set according to the electrochemical parameter set and a preset voltage model.

[0126] Here, the preset voltage model is used to reflect the mapping relationship between electrochemical parameters and voltage. The voltage set includes the voltage of the target battery at different times during multiple charging processes. The preset voltage model uses current density, liquid lithium ion concentration, liquid potential, solid potential and solid lithium ion concentration as input parameters, and voltage as output parameter. Figure 4 The preset voltage model is obtained by pre-training, and the specific process of training the preset voltage model will be described in detail later.

[0127] In some embodiments, the preset voltage model includes one of the following network structure models: convolutional neural network, densely connected network, recurrent neural network, long short-term memory network, attention mechanism model, Transformer, etc.

[0128] Taking the preset voltage model as MLP as an example, the preset voltage model is Figure 4 As shown in m6 in FIG. 1 , the preset voltage model includes an input layer, a hidden layer, and an output layer. The input layer of the preset voltage model is used to input the current density j and the liquid phase lithium ion concentration c. e , liquid phase potential Solid phase potential and the solid phase lithium ion concentration cs The hidden layer of the preset voltage model is used to input the current density j and liquid lithium ion concentration c of the input layer. e , liquid phase potential Solid phase potential and the solid phase lithium ion concentration c s The hidden features are obtained by data processing. The connection matrix between the hidden layer of the preset voltage model and the input layer of the preset voltage model is 5*Q6, where Q6 is the number of neurons in the hidden layer of the preset voltage model, that is, the data dimension of the hidden features output by the hidden layer of the preset voltage model; the output layer of the preset voltage model is used to output voltage U according to the hidden features output by the hidden layer of the preset voltage model, and the connection matrix between the output layer of the preset voltage model and the hidden layer of the preset voltage model is Q6*1. It can be understood that Figure 4 The m6 in the figure is only an example given in this application to explain the preset voltage model and does not limit this application.

[0129] In some embodiments, by Figure 5 The process steps shown, determining the voltage set, include the following steps C1-C6:

[0130] C1. Take the first charging process as the target charging process.

[0131] C2. The first moment in the target charging process is taken as the target moment.

[0132] C3. Input the electrochemical parameters at the target time into a preset voltage model to obtain the voltage at the target time.

[0133] Among them, the current density at the target time, the liquid lithium ion concentration at the target time, the liquid potential at the target time, the solid potential at the target time, and the solid lithium ion concentration at the target time are input into the preset voltage model to obtain the voltage at the target time.

[0134] The preset voltage model is Figure 4 Taking m6 in as an example, the current density at the target moment, the liquid lithium ion concentration at the target moment, the liquid phase potential at the target moment, the solid phase potential at the target moment and the solid phase lithium ion concentration at the target moment are input into the preset voltage model. The current density at the target moment, the liquid lithium ion concentration at the target moment, the liquid phase potential at the target moment, the solid phase potential at the target moment and the solid phase lithium ion concentration at the target moment constitute a 1*5 input matrix. The hidden layer in the preset voltage model multiplies the 1*5 input matrix with the 5*Q6 connection matrix to obtain a 1*Q6 hidden matrix. The output layer in the preset voltage model multiplies the 1*Q6 hidden matrix with the Q6*1 connection matrix to output the voltage at the target moment.

[0135] C4. Determine whether the target time is the last time in the target charging process.

[0136] If the target time is not the last time in the target charging process, execute step C5; if the target time is the last time in the target charging process, execute step C6.

[0137] C5. Set the next moment of the target moment in the target charging process as the target moment, and execute step C3.

[0138] C6. Take the next charging process of the target charging process as the target charging process and execute step C2.

[0139] In the above steps C1-C6, the voltage at a certain moment is obtained by inputting the electrochemical parameters into a preset voltage model that reflects the mapping relationship between the electrochemical parameters and the voltage, and the time and charging process are continuously iterated to determine the voltage at different moments in multiple charging processes.

[0140] S104, generating a full-life charging curve of the target battery according to the voltage set.

[0141] Here, the full life charging curve of the target battery includes the voltage at different times during multiple charging processes. The full life charging curve of the target battery takes time as the horizontal axis and takes the voltage at different times during multiple charging processes as the vertical axis. For example, the full life charging curve of the target battery is as follows: Figure 6 shown.

[0142] It is understandable that the adjacency of two charging processes in the present application refers to the adjacency of the charging processes in sequence rather than the adjacency in charging time. The charging start time of the next charging process of two adjacent charging processes is not necessarily the charging end time of the previous charging process of the two adjacent charging processes.

[0143] In the above Figure 1In the corresponding technical solution, after obtaining a charging plan that reflects the current and temperature of the battery at different times during the charging process, the electrochemical parameter set is determined according to the charging plan and the equation conversion model corresponding to the electrochemical mechanism model, and then the voltage set is determined according to the electrochemical parameter set and the preset voltage model. Finally, the full life curve of the target battery is generated according to the voltage set; since the voltage set includes the voltage of the target battery at different times during multiple charging processes, it is equivalent to predicting the long-term voltage of the battery in the future, and obtaining a charging curve that reflects the long-term battery state in the future is helpful to detect abnormal battery conditions early; since the equation conversion model It is obtained by converting the electrochemical equation in the electrochemical mechanism model. The equation conversion model takes the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter, and takes the dependent variable in the electrochemical equation as an output parameter. The equation conversion model is used to reflect the mapping relationship between the input parameters and the output parameters. It is equivalent to converting the complex electrochemical mechanism model into a simple model that reflects the mapping relationship between the independent variable and the dependent variable in the electrochemical equation to realize the solution of the electrochemical parameters at different times in multiple charging processes, thereby improving the solution speed of the electrochemical parameters at different times in multiple charging processes, thereby improving the speed of generating the full life charging curve of the battery.

[0144] In some embodiments, after generating the full-life charging curve of the target battery, the life duration when the maximum charging capacity of the target battery decays to a preset threshold is determined according to the full-life charging curve of the target battery, and the discharge scheme of the target battery is adjusted based on the life duration of the target battery. The charging capacity of the target battery is obtained based on the voltage corresponding to each charging process on the full-life charging curve of the target battery, and the maximum charging capacity of the target battery is obtained based on the voltage corresponding to the first charging process on the full-life charging curve of the target battery.

[0145] Next, we will introduce the specific process of training and obtaining the equation conversion model corresponding to the electrochemical mechanism model. Figure 7 , Figure 7 The process steps for training and obtaining the equation conversion model corresponding to the electrochemical mechanism model provided in the embodiment of the present application include the following steps S201-S214:

[0146] S201, constructing an electrochemical mechanism model including a thermal field model, and obtaining charging data based on the electrochemical mechanism model simulation.

[0147] The simulation simulates the process of cyclic charging of lithium-ion batteries at different temperatures, different charge and discharge conditions and different aging states, and records the charging data of different groups of lithium-ion batteries during each charging process. The charging data include voltage, current, temperature, current density, solid phase potential, solid phase lithium ion concentration, particle radius, time, liquid phase lithium ion concentration, liquid phase potential during each charging process. The temperature, charge and discharge conditions and aging state corresponding to lithium-ion batteries of the same group are the same, and the temperature and / or charge and discharge conditions and / or aging state corresponding to lithium-ion batteries of different groups are different.

[0148] S202: Acquire a first sample data set from charging data.

[0149] The first sample data set includes a plurality of first sample data, and the first sample data is represented as (j i , ), i represents any time during the charging process, j i represents the current density of the lithium-ion battery at the i-th moment during the charging process, j i is the input sample in the first sample data, represents the liquid phase potential of the lithium-ion battery at the i-th moment during the charging process, is the output sample in the first sample data.

[0150] S203, obtaining a solid phase charge equation conversion model through training according to the first sample data set.

[0151] Among them, the input samples in the first sample data set are input into the solid-phase charge equation conversion model, and the loss of the solid-phase charge equation conversion model is calculated by combining the boundary conditions of the solid-phase charge equation, the output of the solid-phase charge equation conversion model and the output samples corresponding to the input samples in the first sample data set. The parameters of the solid-phase charge equation conversion model are adjusted according to the loss of the solid-phase charge equation conversion model until the solid-phase charge equation conversion model converges, and a solid-phase charge conversion model with the current density in the solid-phase charge equation as the input parameter and the solid-phase potential in the solid-phase charge equation as the output parameter is obtained.

[0152] S204: Acquire a second sample data set from the charging data.

[0153] The second sample data set includes a plurality of second sample data, and the second sample data is represented by (r, t i , c si ), i represents any time during the charging process, r represents the particle radius of the lithium-ion battery, t i represents the time at the i-th moment in the charging process, r and t i is the input sample in the first sample data, c sirepresents the solid phase lithium ion concentration of the lithium ion battery at the i-th moment during the charging process, c si is the output sample in the second sample data.

[0154] S205, obtaining a solid phase mass equation conversion model through training according to the second sample data set.

[0155] Among them, the input samples in the second sample data set are input into the solid-phase mass equation conversion model, and the loss of the solid-phase mass equation conversion model is calculated in combination with the boundary conditions of the solid-phase mass equation, the output of the solid-phase mass equation conversion model and the output samples corresponding to the input samples in the second sample data set. The parameters of the solid-phase mass equation conversion model are adjusted according to the loss of the solid-phase mass equation conversion model until the solid-phase mass equation conversion model converges, and a solid-phase mass equation conversion model is obtained with the particle radius and time in the solid-phase mass equation as input parameters and the solid-phase lithium ion concentration in the solid-phase mass equation as output parameters.

[0156] S206: Acquire a third sample data set from the charging data.

[0157] The third sample data set includes a plurality of third sample data, and the third sample data is represented as (j i , t i+1 , c e(i+1) ), i represents any time during the charging process, j i represents the current density of the lithium-ion battery at the i-th moment during the charging process, t i+1 represents the time of the (i+1)th moment in the charging process, j i and t i+1 is the input sample in the third sample data, c e(i+1) represents the liquid phase lithium ion concentration of the lithium ion battery at the (i+1)th moment during the charging process, c e(i+1) is the output sample in the third sample data.

[0158] S207, obtaining a liquid phase mass equation conversion model by training according to the third sample data set.

[0159] Among them, the input samples in the third sample data set are input into the liquid mass equation conversion model, and the loss of the liquid mass equation conversion model is calculated in combination with the boundary conditions of the liquid mass equation, the output of the liquid mass equation conversion model and the output samples corresponding to the input samples in the third sample data set. The parameters of the liquid mass equation conversion model are adjusted according to the loss of the liquid mass equation conversion model until the liquid mass equation conversion model converges, thereby obtaining a liquid mass equation conversion model with current density and time in the liquid mass equation as input parameters and liquid lithium ion concentration in the liquid mass equation as output parameters.

[0160] S208: Acquire a fourth sample data set from the charging data.

[0161] The fourth sample data set includes a plurality of fourth sample data, and the fourth sample data is represented as (j i , c ei , ), i represents any time during the charging process, j i represents the current density of the lithium-ion battery at the i-th moment during the charging process, c ei represents the liquid phase lithium ion concentration of the lithium ion battery at the i-th moment during the charging process, j i and c ei is the input sample in the fourth sample data, represents the liquid phase potential of the lithium-ion battery at the i-th moment during the charging process, is the output sample in the fourth sample data.

[0162] S209, obtaining a liquid phase charge equation conversion model by training according to the fourth sample data set.

[0163] Among them, the input samples in the fourth sample data set are input into the liquid-phase charge equation conversion model, and the loss of the liquid-phase charge equation conversion model is calculated in combination with the boundary conditions of the liquid-phase charge equation, the output of the liquid-phase charge equation conversion model and the output samples corresponding to the input samples in the fourth sample data set. The parameters of the liquid-phase charge equation conversion model are adjusted according to the loss of the liquid-phase charge equation conversion model until the liquid-phase charge equation conversion model converges, and a liquid-phase charge equation conversion model is obtained with the liquid-phase lithium ion concentration and current density in the liquid-phase charge equation as input parameters and the liquid-phase potential in the liquid-phase charge equation as output parameters.

[0164] S210: Acquire a fifth sample data set from the charging data.

[0165] The fifth sample data set includes a plurality of fifth sample data, and the fifth sample data is represented as (I i , T i , c ei , j i ), i represents any time during the charging process, I i represents the current of the lithium-ion battery at the i-th moment during the charging process, T i represents the temperature of the lithium-ion battery at the i-th moment during the charging process, c ei It represents the liquid phase lithium ion concentration of the lithium ion battery at the i-th moment during the charging process, I i , T i and c ei is the input sample in the fifth sample data, j irepresents the current density of the lithium-ion battery at the i-th moment during the charging process, j i is the output sample in the fifth sample data.

[0166] S211, obtaining a Bavo equation conversion model through training according to the fifth sample data set.

[0167] Among them, the input samples in the fifth sample data set are input into the Bavo equation conversion model, and the loss of the Bavo equation conversion model is calculated by combining the boundary conditions of the Bavo equation, the output of the Bavo equation conversion model and the output samples corresponding to the input samples in the fifth sample data set. The parameters of the Bavo equation conversion model are adjusted according to the loss of the Bavo equation conversion model until the Bavo equation conversion model converges, and a Bavo equation conversion model with the liquid lithium ion concentration, current and temperature in the Bavo equation as input parameters and the current density in the Bavo equation as output parameter is obtained.

[0168] S212, obtaining real voltage data.

[0169] According to various charging conditions in the simulation, the voltage of the lithium-ion battery under various charging conditions is obtained to obtain real voltage data.

[0170] S213, determining a sixth sample data set according to the real voltage data, the solid phase charge equation conversion model, the solid phase mass equation conversion model, the liquid phase mass equation conversion model, the liquid phase charge equation conversion model and the Bavo equation conversion model.

[0171] The sixth sample data set includes a plurality of sixth sample data, and the sixth sample data is represented as ( c sk , c ek , j k , U k ), k represents any time during the charging process, represents the solid phase potential at the kth moment in the charging process output by the solid phase charge equation conversion model, c sk represents the solid phase lithium ion concentration at the kth moment in the charging process output by the solid phase mass equation conversion model, c ek represents the liquid phase lithium ion concentration at the kth moment in the charging process output by the liquid phase mass equation conversion model, represents the liquid potential at the kth moment in the charging process output by the liquid charge equation conversion model, j k represents the current density at the kth moment in the charging process output by the Bavo equation conversion model, c sk , c ek , j kis the input sample in the sixth sample data, U k It represents the voltage of the lithium-ion battery at the kth moment during the charging process, U k is the output sample in the sixth sample data.

[0172] S214: Obtain a preset voltage model through training according to the sixth sample data set.

[0173] Among them, the input samples in the sixth sample data set are input into the preset voltage model, the loss of the preset voltage model is calculated according to the output of the preset voltage model and the output samples corresponding to the input samples in the sixth sample data set, and the parameters of the preset voltage model are adjusted according to the loss of the preset voltage model until the preset voltage model converges to obtain the final preset voltage model.

[0174] In some possible situations, after more charging data of the target battery is acquired, the parameters of the preset voltage model are adjusted according to the acquired charging data.

[0175] The method of the present application is introduced above, and the device of the present application is introduced below.

[0176] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a charging curve generating device provided in an embodiment of the present application, such as Figure 8 As shown, the charging curve generating device 30 includes:

[0177] An acquisition module 301 is used to acquire a charging plan corresponding to a target battery, where the charging plan is used to reflect the current and temperature of the target battery at different times during the charging process;

[0178] A parameter determination module 302, for determining a set of electrochemical parameters according to the aforementioned charging plan and an equation conversion model corresponding to the electrochemical mechanism model;

[0179] The equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model. The equation conversion model uses the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter and uses the dependent variable in the electrochemical equation as an output parameter.

[0180] The aforementioned equation conversion model is used to reflect the mapping relationship between input parameters and output parameters, and the aforementioned electrochemical parameter set includes electrochemical parameters of the aforementioned target battery at different times during multiple charging processes;

[0181] A voltage determination module 303, used to determine a voltage set according to the electrochemical parameter set and a preset voltage model, wherein the preset voltage model is used to reflect the mapping relationship between the electrochemical parameter and the voltage, and the voltage set includes the voltage of the target battery at different times during the multiple charging processes;

[0182] The curve generating module 304 is used to generate a full-life charging curve of the target battery according to the voltage set.

[0183] It should be noted that the charging curve generating device 30 is used to execute the charging curve generating method provided in the embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not fully described in the embodiment of the charging curve generating device 30, please refer to the charging curve generating method provided in the embodiment of the present application.

[0184] See also Fig. 9 , Fig. 9 4 is a schematic diagram of a computer device provided in an embodiment of the present application, wherein the computer device 40 includes a processor 401 and a memory 402. The memory 402 is connected to the processor 401, for example, via a bus.

[0185] The processor 401 is configured to support the computer device 40 to perform the corresponding functions in the method in the above method embodiment. The processor 401 is a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip is an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD is a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0186] The memory 402 is used to store program codes, etc. In some embodiments, the memory 402 includes a volatile memory (VM), such as a random access memory (RAM); in some embodiments, the memory 402 includes a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); in some embodiments, the memory 402 also includes a combination of the above-mentioned types of memory.

[0187] The memory 402 is used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the charging curve generation method in the embodiment of the present application. The processor executes various functional applications and data processing of the charging curve generation method by running the non-volatile software programs, instructions and modules stored in the memory, that is, realizes the functions of the charging curve generation method provided by the above method embodiment.

[0188] In some embodiments, the memory 402 includes a program storage area and a data storage area, wherein the program storage area stores an operating system and an application required for at least one function. The data storage area stores data created according to the use of the charging curve generating device, etc. In some embodiments, the memory includes a memory remotely arranged relative to the processor, and these remote memories are connected to the charging curve generating device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0189] The aforementioned one or more modules are stored in the aforementioned memory, and when executed by the aforementioned one or more processors, the charging curve generation method in any of the aforementioned method embodiments is executed, for example, the method steps described in the above method embodiments are executed to realize the functions of the modules described in the above device embodiments.

[0190] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the method as described in the embodiment.

[0191] Those skilled in the art understand that all or part of the processes in the above-mentioned embodiments are implemented by instructing the relevant hardware through a computer program, and the above-mentioned program is stored in a computer-readable storage medium, and when the program is executed, it includes the processes of the embodiments of the above-mentioned methods. Among them, the above-mentioned storage medium is a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0192] The embodiment of the present application further provides a battery device, which includes the aforementioned charging curve generating device 30 or computer device 40.

[0193] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A charging curve generation method, characterized in that: include: Acquire a charging plan corresponding to a target battery, wherein the charging plan is used to reflect the current and temperature of the target battery at different times during the charging process; Determining a set of electrochemical parameters according to the charging plan and an equation conversion model corresponding to an electrochemical mechanism model; The equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model, and the equation conversion model uses the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter, and uses the dependent variable in the electrochemical equation as an output parameter; The equation conversion model is used to reflect the mapping relationship between input parameters and output parameters, and the electrochemical parameter set includes electrochemical parameters of the target battery at different times during multiple charging processes; Determine a voltage set according to the electrochemical parameter set and a preset voltage model, wherein the preset voltage model is used to reflect a mapping relationship between electrochemical parameters and voltages, and the voltage set includes voltages of the target battery at different times during the multiple charging processes; A full-life charging curve of the target battery is generated according to the voltage set.

2. The method according to claim 1, characterized in that: The electrochemical parameter set includes a current density set, a liquid phase lithium ion concentration set, a liquid phase potential set, a solid phase potential set and a solid phase lithium ion concentration set; The current density set includes the current density of the target battery at different times during the multiple charging processes, the liquid lithium ion concentration set includes the liquid lithium ion concentration of the target battery at different times during the multiple charging processes, the liquid potential set includes the liquid potential of the target battery at different times during the multiple charging processes, the solid potential set includes the solid potential of the target battery at different times during the multiple charging processes, and the solid lithium ion concentration set includes the solid lithium ion concentration of the target battery at different times during the multiple charging processes; The step of determining the electrochemical parameter set according to the equation conversion model corresponding to the charging plan and the electrochemical mechanism model includes: Determining the current density according to the charging plan and the Bavo equation conversion model, wherein the Bavo equation conversion model uses the liquid phase lithium ion concentration, current and temperature in the Bavo equation as input parameters and uses the current density in the Bavo equation as an output parameter; Determining the liquid phase lithium ion concentration according to the current density and the liquid phase mass equation conversion model, wherein the liquid phase mass equation conversion model uses the current density and time in the liquid phase mass equation as input parameters, and uses the liquid phase lithium ion concentration in the liquid phase mass equation as an output parameter; Determining the liquid phase potential according to the current density, the liquid phase lithium ion concentration and the liquid phase charge equation conversion model, wherein the liquid phase charge equation conversion model uses the liquid phase lithium ion concentration and the current density in the liquid phase charge equation as input parameters, and uses the liquid phase potential in the liquid phase charge equation as an output parameter; Determining the solid phase potential according to the current density and the solid phase charge equation conversion model, wherein the solid phase charge equation conversion model uses the current density in the solid phase charge equation as an input parameter and uses the solid phase potential in the solid phase charge equation as an output parameter; The solid phase lithium ion concentration is determined according to the particle radius of the target battery and the solid phase mass equation conversion model. The solid phase mass equation conversion model uses the particle radius and time in the solid phase mass equation as input parameters, and uses the solid phase lithium ion concentration in the solid phase mass equation as an output parameter.

3. The method according to claim 2, characterized in that Determining the current density according to the charging plan and the Bavo equation conversion model includes: Obtaining a current at a target time and a temperature at a target time from the charging plan, wherein the target time is any time during the charging process; The current at the target time, the temperature at the target time, and the liquid phase lithium ion concentration at the target time are input into the Bavo equation conversion model to obtain the current density at the target time.

4. The method according to claim 2, characterized in that: The method of converting the model based on the current density and the liquid mass equation to determine the liquid phase lithium ion concentration includes: The current density at the target moment and the moment next to the target moment are input into the liquid phase mass equation conversion model to obtain the liquid phase lithium ion concentration at the moment next to the target moment, wherein the target moment is any moment in the charging process.

5. The method according to claim 2, characterized in that: The method of determining the liquid phase potential according to the current density, the liquid phase lithium ion concentration and the liquid phase charge equation conversion model comprises: The current density at the target time and the liquid phase lithium ion concentration at the target time are input into the liquid phase charge equation conversion model to obtain the liquid phase potential at the target time, and the target time is any time during the charging process.

6. The method according to claim 2, characterized in that The method of converting the model based on the current density and the solid phase charge equation to determine the solid phase potential comprises: The current density at the target time is input into the solid-phase charge equation conversion model to obtain the solid-phase potential at the target time, and the target time is any time during the charging process.

7. The method according to claim 2, characterized in that: The method of determining the solid phase lithium ion concentration according to the particle radius of the target battery and the solid phase mass equation conversion model comprises: The target time and the particle radius of the target battery are input into the solid phase mass equation conversion model to obtain the solid phase lithium ion concentration at the target time, and the target time is any time during the charging process.

8. The method according to any one of claims 1 to 7, characterized in that: The step of determining a voltage set according to the electrochemical parameter set and a preset voltage model includes: Taking the first charging process as the target charging process, and taking the first moment in the target charging process as the target moment; Inputting the electrochemical parameters at the target time into the preset voltage model to obtain the voltage at the target time; Taking the next moment of the target moment in the target charging process as the target moment, and performing the step of inputting the electrochemical parameters at the target moment into the preset voltage model to obtain the voltage at the target moment, until the target moment is the last moment in the target charging process; The next charging process of the target charging process is taken as the target charging process, and the step of taking the first moment in the target charging process as the target moment is performed.

9. A charging curve generating device, characterized in that: include: An acquisition module, used for acquiring a charging plan corresponding to a target battery, wherein the charging plan is used for reflecting the current and temperature of the target battery at different times during the charging process; a parameter determination module, for determining a set of electrochemical parameters according to the charging plan and an equation conversion model corresponding to the electrochemical mechanism model; The equation conversion model is obtained by converting the electrochemical equation in the electrochemical mechanism model, and the equation conversion model uses the independent variable in the electrochemical equation corresponding to the electrochemical mechanism model as an input parameter, and uses the dependent variable in the electrochemical equation as an output parameter; The equation conversion model is used to reflect the mapping relationship between input parameters and output parameters, and the electrochemical parameter set includes electrochemical parameters of the target battery at different times during multiple charging processes; a voltage determination module, configured to determine a voltage set according to the electrochemical parameter set and a preset voltage model, wherein the preset voltage model is configured to reflect a mapping relationship between electrochemical parameters and voltages, and the voltage set includes voltages of the target battery at different times during the multiple charging processes; A curve generating module is used to generate a full-life charging curve of the target battery according to the voltage set.

10. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1 to 8.

11. A battery device, characterized in that: The charging curve generating device comprises the charging curve generating device as claimed in claim 9, or the computer device as claimed in claim 10.

Images