Battery life prediction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a battery life prediction method and device, computer equipment, a computer readable storage medium and a computer program product, and is applied to the technical field of big data, and the method comprises the steps: obtaining the anode measurement characteristics of a to-be-predicted battery; generating anode characteristics of the to-be-predicted battery according to the anode measurement characteristics of the to-be-predicted battery; and according to the anode characteristics of the to-be-predicted battery, predicting the battery life of the to-be-predicted battery to obtain predicted battery life information. The method can improve the accuracy of battery life prediction.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a battery life prediction method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] With the widespread application and promotion of battery technology, the demand for battery life prediction is increasing for the convenience of use. Therefore, a battery life prediction method is urgently needed.

[0003] In traditional technology, the remaining life of a battery is usually predicted based on the battery usage information. However, the usage information can only indirectly represent the performance of the battery, which can easily result in inaccurate prediction of the remaining life, resulting in low accuracy in battery life prediction. Summary of the invention

[0004] Based on this, it is necessary to provide a battery life prediction method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of battery life prediction in response to the above technical problems.

[0005] In a first aspect, the present application provides a battery life prediction method, comprising:

[0006] Obtaining anode measurement characteristics of the battery to be predicted;

[0007] Generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0008] The battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

[0009] In one of the embodiments, the process of generating the anode characteristics of the battery to be predicted based on the anode measurement characteristics of the battery to be predicted is performed by a pre-trained anode feature extraction model, and the training process of the anode feature extraction model includes: obtaining multiple first training samples, wherein one of the first training samples is generated by the anode characteristics of the training battery; and iteratively training the anode feature extraction model to be trained based on the multiple first training samples to obtain the anode feature extraction model.

[0010] In one embodiment, the anode characteristics of the training battery include preset unit cell strain characteristics and anode property characteristics of the training battery; the obtaining of multiple first training samples includes: obtaining a battery ground state structure of the training battery; simulating the anode property characteristics of the training battery in a battery usage environment corresponding to the preset unit cell strain characteristics according to the battery ground state structure of the training battery to obtain the anode property characteristics of the training battery in the battery usage environment; generating a first training sample according to the anode property characteristics of the training battery in the battery usage environment.

[0011] In one of the embodiments, generating a first training sample based on the anode property characteristics of the training battery in the battery usage environment includes: selecting a first property characteristic from the anode property characteristics of the training battery in the battery usage environment, and determining the first property characteristic as input feature data of the first training sample, and determining a second property characteristic other than the first property characteristic in the anode property characteristics of the training battery in the battery usage environment as a true label corresponding to the input feature data; or, determining the first property characteristic as the input feature data of the first training sample, and jointly determining the second property characteristic and the preset unit cell strain characteristic as the true labels corresponding to the input feature data.

[0012] In one embodiment, the anode property characteristics include structural property characteristics and electrical property characteristics; the anode property characteristics of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics are simulated according to the battery ground state structure of the training battery to obtain the anode property characteristics of the training battery in the battery usage environment, including: simulating the structural changes of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics according to the battery ground state structure of the training battery and the preset unit cell strain characteristics to obtain the structural property characteristics of the training battery in the battery usage environment; simulating the electrical property changes of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics according to the structural property characteristics of the training battery in the battery usage environment to obtain the electrical property characteristics of the training battery in the battery usage environment.

[0013] In one of the embodiments, the anode characteristics of the battery to be predicted include the unit cell strain characteristics and anode property characteristics of the battery to be predicted; predicting the battery life of the battery to be predicted based on the anode characteristics of the battery to be predicted to obtain predicted battery life information includes: obtaining a preset first weight and a preset second weight; weighting the unit cell strain characteristics of the battery to be predicted according to the preset first weight to obtain a strain weighted characteristic, and weighting the anode property characteristics of the battery to be predicted according to the preset second weight to obtain an anode property weighted characteristic; predicting the battery life of the battery to be predicted according to the strain weighted characteristic and the anode property weighted characteristic to obtain predicted battery life information.

[0014] In a second aspect, the present application also provides a battery life prediction device, comprising:

[0015] An acquisition module, used for acquiring anode measurement characteristics of a battery to be predicted;

[0016] A generating module, used for generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0017] The prediction module is used to predict the battery life of the battery to be predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

[0018] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0019] Obtaining anode measurement characteristics of the battery to be predicted;

[0020] Generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0021] The battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

[0022] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0023] Obtaining anode measurement characteristics of the battery to be predicted;

[0024] Generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0025] The battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

[0026] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0027] Obtaining anode measurement characteristics of the battery to be predicted;

[0028] Generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0029] The battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

[0030] The above-mentioned battery life prediction method, device, computer equipment, computer-readable storage medium and computer program product obtain the anode measurement characteristics of the battery to be predicted; generate the anode characteristics of the battery to be predicted based on the anode measurement characteristics of the battery to be predicted; predict the battery life of the battery to be predicted based on the anode characteristics of the battery to be predicted, and obtain predicted battery life information, and use the anode characteristics that can characterize the intrinsic properties of the anode of the battery to be predicted as the basis for battery life prediction, so as to accurately analyze the anode intrinsic performance information of the battery to be predicted, thereby improving the accuracy of battery life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions 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 related drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A diagram showing an application environment of a battery life prediction method in an embodiment;

[0033] Figure 2 is a schematic flow chart of a battery life prediction method in one embodiment;

[0034] Figure 3 A schematic diagram of a flow chart of the steps of anode feature extraction model training process in one embodiment;

[0035] Figure 4 is a structural block diagram of a battery life prediction device in one embodiment;

[0036] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0037] 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.

[0038] It should be noted that the information (including but not limited to anode measurement characteristics, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the acquisition, transmission, storage, use and processing of relevant data are in compliance with the relevant provisions of national laws and regulations. The content pushed to the user (for example, predicted battery life information, etc.) can be rejected by the user or can be easily rejected by the user. In the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0039] The battery life prediction method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the anode measurement characteristics of the battery to be predicted, and generates the anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted; according to the anode characteristics of the battery to be predicted, the battery life of the battery to be predicted is predicted, and the predicted battery life information is obtained, and the predicted battery life information is pushed to the terminal 102. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0040] In an exemplary embodiment, Figure 2 As shown, a battery life prediction method is provided, which is applied to Figure 1The server 104 in the example is used as an example to illustrate the method, which includes the following steps 202 to 206. Among them:

[0041] Step 202: Obtain the anode measurement characteristics of the battery to be predicted.

[0042] Among them, the battery to be predicted in step 202 is a battery waiting for life prediction, and the battery to be predicted can be set by the user as needed. For example, in response to the battery prediction instruction sent by the user terminal, the battery prediction instruction carries the battery information of the battery for which the user performs life prediction as needed. At this time, the battery to be predicted is the battery corresponding to the battery information carried in the battery prediction instruction. In this way, the battery life prediction can be performed according to user needs; it can also be determined by satisfying preset conditions. The preset condition can be that the battery usage time exceeds the preset time threshold. At this time, the battery to be predicted is a battery whose corresponding battery usage time exceeds the preset time threshold; it can also be that the number of battery cycles is greater than the preset number of cycle threshold. At this time, the battery to be predicted is a battery whose corresponding battery cycle number is greater than the preset number of cycle threshold. In this way, in some scenarios where the scale of battery use is large, such as factories, etc., the life prediction of batteries with a low risk of remaining life can be carried out in time, which can facilitate subsequent battery repair or replacement, thereby avoiding unnecessary losses due to battery aging, damage, etc.

[0043] Among them, the battery to be predicted in step 202 can be a hydrogen fuel cell or a lithium battery, which is not limited here. A hydrogen fuel cell is a power generation device that uses hydrogen stored in a hydrogen storage medium to generate electrical energy, and has the advantages of high efficiency and zero carbon emissions. The current focus of hydrogen fuel cells is on improving the relevant performance of their key components, and the anode feature generation in this scheme can help researchers understand and optimize the working principle and performance of the battery, reduce the test and experimental cycle, and save R&D costs and resource investment. Hydrogen fuel cells include but are not limited to alkaline hydrogen fuel cells (Alkaline Fuel Cell, AFC), proton exchange membrane hydrogen fuel cells (Proton Exchange Membrane FuelCell, PEMFC), phosphoric acid hydrogen fuel cells (Phosphoric Acid Fuel Cell, PAFC), molten carbonate hydrogen fuel cells (Molten Carbonate Fuel Cell, MCFC), solid oxide hydrogen fuel cells (Solid Oxide FuelCell, SOFC), etc.

[0044] Wherein, when the battery in step 202 is a hydrogen fuel cell, the anode of the battery includes but is not limited to carbon materials, precious metal materials, nickel-based metal materials, yttria-stabilized zirconia (YSZ) materials, and the like.

[0045] The anode measurement characteristics in step 202 are parameters measured by a measuring device and used to characterize the battery anode performance of the battery to be predicted. The number of parameters of the anode measurement characteristics can be one or more, and there is no limitation here. The anode measurement characteristics include electrical measurement property characteristics. The electrical measurement property characteristics include at least one of electronic property characteristics, magnetic property characteristics, electrochemical property characteristics and ion diffusion property characteristics; the electronic property characteristics include at least one of cutoff energy, Brillouin zone k-point, Gaussian broadening, relaxation convergence accuracy, interatomic interaction force, ISMEAR and SIGMA; the magnetic property characteristics include at least one of cutoff energy, Brillouin zone k-point, Gaussian broadening, relaxation convergence accuracy, interatomic interaction force, spin polarization and magnetic moment of each atom; the electrochemical property characteristics include at least one of cutoff energy, Brillouin zone k-point, Gaussian broadening, relaxation convergence accuracy, interatomic interaction force, LEPSILON and LOPTICS; the ion diffusion property characteristics include at least one of cutoff energy, Brillouin zone k-point, Gaussian broadening, relaxation convergence accuracy, interatomic interaction force, ensemble, thermostat, step size of atomic movement, number of steps of atomic movement, initial temperature and kinetic step size.

[0046] Exemplarily, step 202 includes: acquiring anode measurement characteristics of the battery to be predicted collected by a sensor device.

[0047] It is understandable that since the anode measurement characteristics are the basis for generating the anode characteristics, it is necessary to ensure that the collected anode measurement characteristics are easy to collect and / or accurate, thereby ensuring the accuracy of the subsequent generation of the anode characteristics of the battery to be predicted.

[0048] As an embodiment, step 202 includes: obtaining measurement condition information, selecting a collection data type from electronic property characteristics according to the measurement condition information; and obtaining anode measurement characteristics of the battery to be predicted according to the collection data type.

[0049] In this way, it can be ensured that the collected anode measurement characteristics are easy to collect and accurate under the measurement environment corresponding to the current measurement condition information, thereby ensuring the accuracy of the anode characteristics of the battery to be predicted that are subsequently generated.

[0050] Step 204 , generating anode characteristics of the battery to be predicted based on the measured anode characteristics of the battery to be predicted.

[0051] Exemplarily, step 204 includes: mapping the anode measurement features of the battery to be predicted into the anode features of the battery to be predicted by using an anode feature extraction model. The anode feature extraction model may be a neural network model or a mathematical model.

[0052] As an embodiment, the anode measurement characteristics of the battery to be predicted are mapped to the anode characteristics of the battery to be predicted through an anode feature extraction model, including: when the anode feature extraction model is a mathematical model, the anode feature extraction model includes a structural property feature extraction model and an electrical property feature extraction model, the anode measurement characteristics of the battery to be predicted are mapped to the structural property characteristics of the battery to be predicted through the structural property feature extraction model, the structural property characteristics of the battery to be predicted are mapped to the electrical property characteristics of the battery to be predicted through the electrical property feature extraction model, and the structural property characteristics and electrical property characteristics of the battery to be predicted are fused to obtain the anode characteristics of the battery to be predicted.

[0053] As an embodiment, the anode measurement features of the battery to be predicted are mapped to the anode features of the battery to be predicted through an anode feature extraction model, including: when the anode feature extraction model is a neural network model, the anode features include features in the anode property features except the anode measurement features, and the anode measurement features of the battery to be predicted are mapped to features in the anode property features of the battery to be predicted except the anode measurement features through the anode feature extraction model.

[0054] As another embodiment, the anode measurement characteristics of the battery to be predicted are mapped to the anode characteristics of the battery to be predicted through the anode feature extraction model, including: when the anode feature extraction model is a neural network model, the anode characteristics include characteristics of the anode property characteristics other than the anode measurement characteristics, structural property characteristics and unit cell strain characteristics, and the anode measurement characteristics of the battery to be predicted are mapped to the characteristics of the anode property characteristics of the battery to be predicted other than the anode measurement characteristics and the unit cell strain characteristics through the anode feature extraction model.

[0055] Step 206 , predicting the battery life of the battery to be predicted based on the anode characteristics of the battery to be predicted, and obtaining predicted battery life information.

[0056] The predicted battery life information in step 206 is used to characterize the remaining battery life of the battery to be predicted. The predicted battery life information may be the predicted remaining battery usage time or the predicted remaining battery cycle count.

[0057] Exemplarily, step 206 includes: mapping the anode characteristics of the battery to be predicted into predicted battery life information of the battery to be predicted by using a battery life prediction model.

[0058] As an embodiment, step 206 includes: the anode characteristics of the battery to be predicted include the structural property characteristics and electrical property characteristics of the battery to be predicted; the structural property characteristics and electrical property characteristics of the battery to be predicted are spliced ​​together through a battery life prediction model to obtain the battery characteristics of the battery to be predicted; and the battery characteristics of the battery to be predicted are mapped into predicted battery life information of the battery to be predicted through the battery life prediction model.

[0059] As another embodiment, step 206 includes: the anode characteristics of the battery to be predicted include characteristics other than the anode measurement characteristics and the unit cell strain characteristics in the anode property characteristics of the battery to be predicted; the characteristics other than the anode measurement characteristics, the anode measurement characteristics and the unit cell strain characteristics in the anode property characteristics of the battery to be predicted are spliced ​​through a battery life prediction model to obtain battery characteristics of the battery to be predicted; and the battery characteristics of the battery to be predicted are mapped into predicted battery life information of the battery to be predicted through the battery life prediction model.

[0060] It is understandable that during the use of the battery, the cell structure of the battery may undergo tensile strain and / or contraction strain, which will affect the intrinsic properties of the battery anode. In this way, the cell strain characteristics are used as one of the bases for predicting the life of the battery to be predicted, and the influence of the strain of the cell structure of the battery during the use of the battery on the battery life is taken into account, which improves the accuracy of the battery life prediction.

[0061] As an embodiment, step 206 includes: obtaining a preset first weight and a preset second weight; weighting the unit cell strain characteristics of the battery to be predicted according to the preset first weight to obtain a strain weighted characteristic, and weighting the anode property characteristics of the battery to be predicted according to the preset second weight to obtain an anode property weighted characteristic; predicting the battery life of the battery to be predicted according to the strain weighted characteristic and the anode property weighted characteristic to obtain predicted battery life information.

[0062] Among them, the preset first weight and the preset second weight can be set by the user as needed, and can also be set according to experience. If the unit cell strain characteristics have a greater impact on the accuracy of the battery life prediction of the battery to be predicted than the anode property characteristics, the preset first weight will be set larger than the preset second weight. If the unit cell strain characteristics have a smaller impact on the accuracy of the battery life prediction of the battery to be predicted than the anode property characteristics, the preset first weight will be set smaller than the preset second weight.

[0063] Among them, the process of predicting the battery life of the battery to be predicted and obtaining the predicted battery life information can be performed by a pre-trained battery life prediction model. The training process of the battery life prediction model can refer to the training process of the anode feature extraction model and be adaptively adjusted, which will not be elaborated here.

[0064] Optionally, the model type of the battery life prediction model includes but is not limited to a multivariate linear regression model type, a support vector machine model type, a decision tree regression model type, a neural network model type, and the like.

[0065] It is understandable that, since the degradation process of the anode performance during the use of the battery is usually nonlinear and affected by a variety of complex factors, when the model type of the battery life prediction model adopts the decision tree regression model type, the battery life prediction model can divide the anode characteristics into different intervals and perform local linear fitting in each interval to capture these nonlinear relationships, thereby effectively predicting the remaining life of the battery to be predicted; and at this time, the battery life prediction model can automatically select the most useful features for prediction during the construction process, and can effectively identify the parameters that have the greatest impact on the battery life, which enables the battery life prediction model to focus on key variables, thereby improving the accuracy of battery life prediction. In addition, battery life prediction involves multiple types of data, that is, the anode characteristics are high-dimensional data sets. When the model type of the battery life prediction model adopts the decision tree regression model type, the battery life prediction model can effectively process the anode characteristics and give accurate battery life predictions by splitting the data layer by layer and gradually narrowing the prediction range; the battery life prediction model can also process time series data through feature engineering (such as converting time series data into feature variables) to predict the future life decay trend of the battery. Therefore, the decision tree regression model type is preferably used as the model type of the battery life prediction model.

[0066] In the above-mentioned battery life prediction method, the anode measurement characteristics of the battery to be predicted are obtained; the anode characteristics of the battery to be predicted are generated according to the anode measurement characteristics of the battery to be predicted; the battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, and the predicted battery life information is obtained. The anode characteristics that can characterize the intrinsic properties of the anode of the battery to be predicted are used as the basis for battery life prediction, so that the intrinsic performance information of the anode of the battery to be predicted can be accurately analyzed, thereby improving the accuracy of battery life prediction.

[0067] In an exemplary embodiment, Figure 3 As shown, the process of generating the anode features of the battery to be predicted according to the anode measurement features of the battery to be predicted is performed by a pre-trained anode feature extraction model, and the training process of the anode feature extraction model includes steps 302 to 304. Among them:

[0068] Step 302: Acquire multiple first training samples, wherein one first training sample is generated by anode characteristics of a training battery.

[0069] Exemplarily, step 302 includes: obtaining anode characteristics of a training battery, and generating a plurality of first training samples according to the anode characteristics of the training battery.

[0070] As an embodiment, step 302 includes: the anode characteristics of the training battery include preset unit cell strain characteristics and anode property characteristics of the training battery; obtaining the battery ground state structure of the training battery; simulating the anode property characteristics of the training battery in a battery usage environment corresponding to the preset unit cell strain characteristics based on the battery ground state structure of the training battery to obtain the anode property characteristics of the training battery in the battery usage environment; generating a first training sample based on the anode property characteristics of the training battery in the battery usage environment.

[0071] Among them, the anode property characteristics include electrical property characteristics, and the electrical property characteristics include at least one of electronic property characteristics, magnetic property characteristics, electrochemical property characteristics and ion diffusion property characteristics; the electronic property characteristics, magnetic property characteristics, electrochemical property characteristics and ion diffusion property characteristics can refer to the further explanation content in step 202, and will not be repeated here.

[0072] Among them, the preset unit cell strain feature is a pre-set strain feature for applying strain to the unit cells of the training battery. The strain application type of the preset unit cell strain feature can be a compression strain type or a contraction strain type. After the preset strain is applied to the unit cells of the training battery, a battery usage environment is simulated.

[0073] As an embodiment, the preset unit cell strain characteristics include a preset unit cell strain amount, which is used to characterize the change in the unit cell constant of the unit cell structure of the training battery. The preset unit cell strain amount is determined by the intrinsic lattice constant and the change in the lattice constant of the unit cell structure of the training battery. Specifically, the preset unit cell strain amount is the difference between the intrinsic lattice constant and the change in the lattice constant, and the ratio between the intrinsic lattice constant.

[0074] Optionally, the preset unit cell strain is the ratio of the difference between the intrinsic lattice constant and the lattice constant variation to the intrinsic lattice constant, which can be expressed by the formula:

[0075]

[0076] in, is the preset unit cell strain, is the intrinsic lattice constant, is the change in lattice constant.

[0077] Furthermore, obtaining the battery ground state structure of the training battery includes: obtaining a unit cell structure model of the training battery, and extracting the battery ground state structure of the training battery from the unit cell structure model.

[0078] As an embodiment, obtaining a unit cell structure model of a training battery includes: obtaining a battery model of the training battery, and querying the unit cell structure model of the training battery from a preset unit cell model library according to the battery model of the training battery, wherein the preset unit cell model library includes a correspondence between the battery model and the unit cell structure model of the battery.

[0079] As an embodiment, according to the battery ground state structure of the training battery, the anode property characteristics of the training battery in a battery usage environment corresponding to the preset unit cell strain characteristics are simulated to obtain the anode property characteristics of the training battery in the battery usage environment, including: the anode property characteristics include structural property characteristics and electrical property characteristics; according to the battery ground state structure and the preset unit cell strain characteristics of the training battery, the structural changes of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics are simulated to obtain the structural property characteristics of the training battery in the battery usage environment; according to the structural property characteristics of the training battery in the battery usage environment, the electrical property changes of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics are simulated to obtain the electrical property characteristics of the training battery in the battery usage environment.

[0080] Among them, the simulation of the structural changes and electrical property changes of the training battery under the battery use environment can be simulated based on the density functional theory, which includes but is not limited to the Born-Oppenheimer approximation, Hartree-Fock ( ) approximation and Kohn-Sham equation can also be simulated based on Newtonian mechanics theory.

[0081] Optionally, when using Newtonian mechanics theory to simulate the structural changes and electrical property changes of the training battery in the battery usage environment, it is necessary to measure more calculation data, while when using density functional theory to simulate the structural changes and electrical property changes of the training battery in the battery usage environment, it is not necessary to measure the calculation data, and the simulation calculation of the structural changes and electrical property changes can be achieved with the battery ground state structure. Therefore, it is preferred to use density functional theory to simulate the structural changes and electrical property changes of the training battery in the battery usage environment.

[0082] As one embodiment, a first training sample is generated based on the anode property characteristics of a training battery in a battery usage environment, including: selecting a first property characteristic from the anode property characteristics of the training battery in the battery usage environment, and determining the first property characteristic as input feature data of the first training sample, and determining a second property characteristic other than the first property characteristic in the anode property characteristics of the training battery in the battery usage environment as a true label corresponding to the input feature data; or, determining the first property characteristic as the input feature data of the first training sample, and jointly determining the second property characteristic and a preset unit cell strain characteristic as the true labels corresponding to the input feature data.

[0083] Step 304: iteratively train the anode feature extraction model to be trained according to the plurality of first training samples to obtain the anode feature extraction model.

[0084] Exemplarily, step 304 includes: each first training sample consists of input feature data and a true label corresponding to the input feature data, for each training sample, mapping the input feature data in the first training sample to a first training label through the anode feature extraction model to be trained, and calculating the model loss corresponding to the anode feature extraction model to be trained according to the difference between the first training label and the true label in the first training sample; judging whether the model loss corresponding to the anode feature extraction model to be trained converges, if the model loss corresponding to the anode feature extraction model to be trained converges, taking the current anode feature extraction model to be trained as the trained anode feature extraction model; if the model loss corresponding to the anode feature extraction model to be trained does not converge, updating the anode feature extraction model to be trained according to the gradient calculated by the model loss corresponding to the anode feature extraction model to be trained through a preset model updating method, and returning to the step of mapping the input feature data in the first training sample to a first training label through the anode feature extraction model to be trained, until the model loss corresponding to the anode feature extraction model to be trained converges, wherein the preset model updating method includes a gradient descent method and a gradient ascent method, etc.

[0085] As an embodiment, when the second property characteristic other than the first property characteristic in the anode property characteristic of the training battery under the battery usage environment is determined as the true label corresponding to the input feature data, the first training label includes the training second property characteristic other than the first property characteristic in the anode property characteristic of the training battery under the battery usage environment; or, when the second property characteristic and the preset unit cell strain characteristic are jointly determined as the true label corresponding to the input feature data, the first training label includes the training second property characteristic and the training unit cell strain characteristic.

[0086] Optionally, before performing model training, the method also includes: adding Gaussian noise with a preset standard deviation to the first training sample, wherein the preset standard deviation can be 0.01 or other set values, which are not limited here, and wherein Gaussian noise is random noise that obeys a normal distribution.

[0087] In this way, the generalization of the trained anode feature extraction model can be improved, that is, the anode feature prediction accuracy of the anode feature extraction model can be improved.

[0088] Optionally, before performing model training, the method also includes: randomly dividing the first training samples (to which Gaussian noise with a preset standard deviation has been added) into a training set and a test set according to a preset ratio, and training the anode feature extraction model to be trained based on the training set and the test set, wherein the preset ratio can be 4:1 or other ratios, which are not limited here.

[0089] As an embodiment, after the anode feature extraction model to be trained is trained according to the training set and the test set, it also includes: calculating a model evaluation index of the anode feature extraction model, and evaluating the model effect of the anode feature extraction model according to the model evaluation index.

[0090] Calculating the model evaluation index of the anode feature extraction model includes: calculating the mean square error and the determination coefficient of the anode feature extraction model according to the differences between each first training label and the real label in the first training sample, and the sample number of the first training sample, and determining the mean square error and the determination coefficient of the anode feature extraction model as the model evaluation index of the anode feature extraction model.

[0091] Furthermore, the mean square error of the anode feature extraction model is calculated based on the differences between each first training label and the true label in the first training sample, as well as the sample number of the first training sample, including: based on the sample number of the first training sample, determining the average of the squares of the differences between each first training label and the true label in the first training sample as the mean square error of the anode feature extraction model.

[0092] Optionally, according to the number of samples of the first training sample, the average value of the square of the difference between each first training label and the real label in the first training sample is determined as the mean square error of the anode feature extraction model, which can be expressed by the formula:

[0093]

[0094] in, is the mean square error of the anode feature extraction model, is the number of samples of the first training sample, For the The true label of the first training sample, For the The first training label corresponding to the first training sample.

[0095] Furthermore, based on the differences between each first training label and the true label in the first training sample, as well as the sample number of the first training sample, the determination coefficient of the anode feature extraction model is calculated, including: determining the sum of the squares of the differences between each first training label and the true label in the first training sample as the first parameter, determining the sum of the squares of the differences between the true label in each first training sample and the average value of the first training labels as the second parameter, determining the ratio between the first parameter and the second parameter as the third parameter, and determining the difference between the preset parameter and the third parameter as the determination coefficient of the anode feature extraction model, wherein the preset parameter can be 1.

[0096] Optionally, when the preset parameter is 1, according to the difference between each first training label and the real label in the first training sample, and the number of samples of the first training sample, the determination coefficient of the anode feature extraction model can be expressed by the formula:

[0097]

[0098] in, is the determination coefficient of the anode feature extraction model, is the average of the first training labels.

[0099] As an embodiment, the model effect of the anode feature extraction model is evaluated according to the model evaluation index, including: when the model evaluation index includes the mean square error of the anode feature extraction model, the smaller the mean square error of the anode feature extraction model is, the better the model effect of the anode feature extraction model obtained by evaluation is; when the model evaluation index includes the determination coefficient of the anode feature extraction model, the larger the determination coefficient of the anode feature extraction model is, the better the model effect of the anode feature extraction model obtained by evaluation is.

[0100] In this embodiment, a plurality of first training samples are obtained, wherein one first training sample is generated by the anode feature of a training battery; based on the plurality of first training samples, an anode feature extraction model to be trained is iteratively trained to obtain an anode feature extraction model, and a training process for an anode feature extraction model is provided, so that a model for predicting battery anode features can be trained, thereby improving the accuracy of battery anode feature prediction.

[0101] As a detailed embodiment, the anode characteristics of the training battery include preset unit cell strain characteristics and anode property characteristics of the training battery; the battery ground state structure of the training battery is obtained; the anode property characteristics include structural property characteristics and electrical property characteristics; according to the battery ground state structure and the preset unit cell strain characteristics of the training battery, the structural changes of the training battery in the battery use environment corresponding to the preset unit cell strain characteristics are simulated to obtain the structural property characteristics of the training battery in the battery use environment; according to the structural property characteristics of the training battery in the battery use environment, the electrical property changes of the training battery in the battery use environment corresponding to the preset unit cell strain characteristics are simulated to obtain the training battery in the battery Electrical property characteristics under the battery usage environment; selecting a first property characteristic from the anode property characteristics of the training battery under the battery usage environment, and determining the first property characteristic as the input feature data of the first training sample, and determining the second property characteristic other than the first property characteristic in the anode property characteristics of the training battery under the battery usage environment as the true label corresponding to the input feature data; or, determining the first property characteristic as the input feature data of the first training sample, and jointly determining the second property characteristic and the preset unit cell strain characteristic as the true label corresponding to the input feature data; according to multiple first training samples, iteratively training the anode feature extraction model to be trained to obtain the anode feature extraction model.

[0102] Furthermore, the anode measurement characteristics of the battery to be predicted are obtained; the anode characteristics of the battery to be predicted are generated according to the anode measurement characteristics of the battery to be predicted by a pre-trained anode feature extraction model; the anode characteristics of the battery to be predicted include the unit cell strain characteristics and the anode property characteristics of the battery to be predicted; a preset first weight and a preset second weight are obtained; according to the preset first weight, the unit cell strain characteristics of the battery to be predicted are weighted to obtain strain weighted characteristics, and, according to the preset second weight, the anode property characteristics of the battery to be predicted are weighted to obtain anode property weighted characteristics; according to the strain weighted characteristics and the anode property weighted characteristics, the battery life of the battery to be predicted is predicted to obtain predicted battery life information.

[0103] In this way, by obtaining the anode measurement characteristics of the battery to be predicted; generating the anode characteristics of the battery to be predicted based on the anode measurement characteristics of the battery to be predicted; predicting the battery life of the battery to be predicted based on the anode characteristics of the battery to be predicted, and obtaining the predicted battery life information, the anode characteristics that can characterize the intrinsic properties of the anode of the battery to be predicted are used as the basis for battery life prediction, so that the anode intrinsic performance information of the battery to be predicted can be accurately analyzed, thereby improving the accuracy of battery life prediction.

[0104] Furthermore, by obtaining multiple first training samples, wherein one first training sample is generated by the anode feature of the training battery; based on the multiple first training samples, the anode feature extraction model to be trained is iteratively trained to obtain the anode feature extraction model, and a training process for the anode feature extraction model is provided, which can train a model for predicting the battery anode characteristics, thereby improving the accuracy of the battery anode feature prediction.

[0105] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0106] Based on the same inventive concept, the embodiment of the present application also provides a battery life prediction device for implementing the battery life prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more battery life prediction device embodiments provided below can refer to the limitations of the battery life prediction method above, and will not be repeated here.

[0107] In an exemplary embodiment, Figure 4 As shown, a battery life prediction device 400 is provided, comprising: an acquisition module 402, a generation module 404 and a prediction module 406, wherein:

[0108] An acquisition module 402 is used to acquire anode measurement characteristics of a battery to be predicted;

[0109] A generating module 404, configured to generate anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted;

[0110] The prediction module 406 is used to predict the battery life of the battery to be predicted according to the anode characteristics of the battery to be predicted, and obtain predicted battery life information.

[0111] In one of the embodiments, the process of generating the anode characteristics of the battery to be predicted is performed by a pre-trained anode feature extraction model based on the anode measurement characteristics of the battery to be predicted; the device also includes: a training module for acquiring multiple first training samples, wherein one first training sample is generated by the anode characteristics of the training battery; based on the multiple first training samples, the anode feature extraction model to be trained is iteratively trained to obtain the anode feature extraction model.

[0112] In one of the embodiments, the anode characteristics of the training battery include preset unit cell strain characteristics and anode property characteristics of the training battery; the training module is also used to obtain the battery ground state structure of the training battery; based on the battery ground state structure of the training battery, the anode property characteristics of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics are simulated to obtain the anode property characteristics of the training battery in the battery usage environment; based on the anode property characteristics of the training battery in the battery usage environment, a first training sample is generated.

[0113] In one of the embodiments, the training module is also used to select a first property characteristic from the anode property characteristics of the training battery under the battery usage environment, and determine the first property characteristic as the input feature data of the first training sample, and determine the second property characteristic other than the first property characteristic in the anode property characteristics of the training battery under the battery usage environment as the true label corresponding to the input feature data; or, determine the first property characteristic as the input feature data of the first training sample, and determine the second property characteristic and the preset unit cell strain characteristic together as the true label corresponding to the input feature data.

[0114] In one of the embodiments, the anode property characteristics include structural property characteristics and electrical property characteristics; the training module is further used to simulate the structural changes of the training battery in a battery usage environment corresponding to the preset unit cell strain characteristics based on the battery ground state structure and preset unit cell strain characteristics of the training battery, so as to obtain the structural property characteristics of the training battery in the battery usage environment; based on the structural property characteristics of the training battery in the battery usage environment, simulate the electrical property changes of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics, so as to obtain the electrical property characteristics of the training battery in the battery usage environment.

[0115] In one of the embodiments, the anode characteristics of the battery to be predicted include the unit cell strain characteristics and the anode property characteristics of the battery to be predicted; the prediction module 406 is also used to obtain a preset first weight and a preset second weight; according to the preset first weight, the unit cell strain characteristics of the battery to be predicted are weighted to obtain a strain weighted characteristic, and, according to the preset second weight, the anode property characteristics of the battery to be predicted are weighted to obtain anode property weighted characteristics; based on the strain weighted characteristics and the anode property weighted characteristics, the battery life of the battery to be predicted is predicted to obtain predicted battery life information.

[0116] Each module in the above-mentioned battery life prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0117] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a battery life prediction method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0118] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0120] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0121] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0122] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0124] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A battery life prediction method, characterized in that: The method comprises: Obtaining anode measurement characteristics of the battery to be predicted; Generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted; The battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

2. The method according to claim 1, characterized in that The process of generating the anode feature of the battery to be predicted according to the anode measurement feature of the battery to be predicted is performed by a pre-trained anode feature extraction model; the training process of the anode feature extraction model includes: Acquire a plurality of first training samples, wherein one of the first training samples is generated by an anode feature of a training battery; The anode feature extraction model to be trained is iteratively trained according to the multiple first training samples to obtain the anode feature extraction model.

3. The method according to claim 2, characterized in that The anode characteristics of the training battery include preset unit cell strain characteristics and anode property characteristics of the training battery; The obtaining of a plurality of first training samples comprises: Get the battery ground state structure of the training battery; According to the battery ground state structure of the training battery, simulating the anode property characteristics of the training battery in the battery use environment corresponding to the preset unit cell strain characteristics, to obtain the anode property characteristics of the training battery in the battery use environment; A first training sample is generated according to the anode property characteristics of the training battery under the battery usage environment.

4. The method according to claim 3, characterized in that The step of generating a first training sample according to the anode property characteristics of the training battery in the battery use environment includes: Selecting a first property feature from the anode property features of the training battery in the battery use environment, and determining the first property feature as input feature data of a first training sample, and determining a second property feature other than the first property feature among the anode property features of the training battery in the battery use environment as a true label corresponding to the input feature data; or, The first property feature is determined as input feature data of the first training sample, and the second property feature and the preset unit cell strain feature are jointly determined as true labels corresponding to the input feature data.

5. The method according to claim 3, characterized in that: The anode property characteristics include structural property characteristics and electrical property characteristics; the anode property characteristics of the training battery in the battery use environment corresponding to the preset unit cell strain characteristics are simulated according to the battery ground state structure of the training battery to obtain the anode property characteristics of the training battery in the battery use environment, including: According to the battery ground state structure of the training battery and the preset unit cell strain characteristics, the structural change of the training battery under the battery use environment corresponding to the preset unit cell strain characteristics is simulated to obtain the structural property characteristics of the training battery under the battery use environment; According to the structural property characteristics of the training battery in the battery usage environment, the change of the electrical properties of the training battery in the battery usage environment corresponding to the preset unit cell strain characteristics is simulated to obtain the electrical property characteristics of the training battery in the battery usage environment.

6. The method according to claim 1, characterized in that The anode characteristics of the battery to be predicted include the unit cell strain characteristics and anode property characteristics of the battery to be predicted; and the battery life of the battery to be predicted is predicted according to the anode characteristics of the battery to be predicted to obtain predicted battery life information, including: Obtaining a preset first weight and a preset second weight; According to the preset first weight, weighted processing is performed on the unit cell strain characteristics of the battery to be predicted to obtain a strain weighted characteristic, and, according to the preset second weight, weighted processing is performed on the anode property characteristics of the battery to be predicted to obtain an anode property weighted characteristic; The battery life of the battery to be predicted is predicted according to the strain weighted feature and the anode property weighted feature to obtain predicted battery life information.

7. A battery life prediction device, characterized in that: The device comprises: An acquisition module, used for acquiring anode measurement characteristics of a battery to be predicted; A generating module, used for generating anode characteristics of the battery to be predicted according to the anode measurement characteristics of the battery to be predicted; The prediction module is used to predict the battery life of the battery to be predicted according to the anode characteristics of the battery to be predicted, so as to obtain predicted battery life information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images