Method, device, electronic terminal, and storage medium for predicting battery charge and discharge cycle life

By extracting the early charging and discharging characteristics of lithium batteries and using clustering and regression analysis algorithms, a model is established to predict the charging and discharging cycle life, solving the problem of insufficient prediction accuracy in the existing technology, and achieving high-efficiency life prediction for different types of batteries.

CN112084459BActive Publication Date: 2024-12-03DENKEI TRADING (SHANGHAI) CO LTD
View PDF 14 Cites 0 Cited by

Patent Information

Application Number
CN201910506083.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-06-12
Publication Date
2024-12-03
Estimated Expiration
2040-05-31

AI Technical Summary

Technical Problem

The prior art is difficult to accurately predict the charge and discharge cycle life of lithium batteries, especially in the absence of existing life attenuation data and a large amount of historical data required, resulting in inaccurate prediction results.

Method used

By obtaining the early charging and discharging cycle data of the battery, extracting the charging and discharging characteristics, and using clustering analysis and regression analysis algorithms to build models to predict the charging and discharging cycle life, only early data is required and does not rely on complex physical and chemical mechanisms.

Benefits of technology

It reduces software and hardware requirements, simplifies the deployment of prediction systems, and can more accurately predict the charging and discharge cycle life of different types of batteries, improving the convenience and accuracy of prediction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112084459B_ABST
    Figure CN112084459B_ABST
Patent Text Reader

Abstract

The present application provides a method, device, electronic terminal, and storage medium for predicting battery cycle life, which includes: obtaining early charge and discharge cycle data of a battery set; extracting charge and discharge cycle characteristics of at least one battery set based on the obtained early charge and discharge cycle data, and generating a charge and discharge cycle data feature vector of the battery set to form a corresponding charge and discharge cycle feature vector set; clustering and grouping the charge and discharge cycle data feature vector set using a clustering analysis algorithm; and establishing corresponding regression models for one or more groups of clustered charge and discharge cycle data feature vector subsets using a regression analysis algorithm, so as to predict the charge and discharge cycle life of the battery according to a regression model that is closest to the characteristics of the battery to be tested in the established regression model. The present application can be easily extended to the prediction of charge and discharge cycle life of different types of batteries, reduce the software and hardware requirements of the prediction system, and facilitate deployment and use in practical applications.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of battery charge and discharge cycle life prediction, and in particular to a battery charge and discharge cycle life prediction method, device, electronic terminal, and storage medium. Background Art

[0002] With the popularization and promotion of new energy vehicles, the charge and discharge cycle life prediction of power batteries, one of the core components of new energy vehicles, is increasingly attracting the attention and importance of all walks of life in the industry, including power battery manufacturers, automobile manufacturers and end users. For lithium batteries, when their actual capacity drops to 80% of their rated capacity, they are considered to be invalid. Long charge and discharge cycle life will bring huge market competitiveness to new energy vehicles, so it is of great significance to predict the charge and discharge cycle life of lithium batteries.

[0003] Existing data-driven prediction technologies generally use a particle filter algorithm to predict the service life of a battery. The initial values ​​of the parameters of the particle filter model can usually be obtained through experience or calculated through known life decay data of other batteries.

[0004] However, in actual application, due to the complex physical and chemical processes inside lithium batteries, if the initial values ​​of the parameters need to be accurately estimated, high requirements are often placed on the user's experience. For newly designed batteries, there may be no life attenuation data of existing batteries that can be used as a reference. At the same time, in the prediction process, the existing technology often requires all historical charge and discharge cycle data of the prediction object in order to update the battery's own status. In these cases, the parameters obtained often cannot truly reflect the actual conditions of the lithium battery, resulting in inaccurate prediction results for the final charge and discharge cycle life.

[0005] At the same time, for the prediction object, the existing technology often needs to use as much historical charge and discharge cycle data as possible, usually requiring a large amount of historical data of batteries, and cannot achieve accurate prediction of cycle life through partial or early charge and discharge cycle data.

[0006] Application Contents

[0007] In view of the shortcomings of the prior art described above, the purpose of the present application is to provide a method, device, electronic terminal, and storage medium for predicting the battery charge and discharge cycle life, so as to solve the technical problem in the prior art that it is difficult to obtain the initial values ​​of parameters used to predict the battery charge and discharge cycle life.

[0008] To achieve the above-mentioned purpose and other related purposes, the first aspect of the present application provides a method for predicting the charge and discharge cycle life of a battery, which includes: obtaining early charge and discharge cycle data of a battery set; extracting the charge and discharge characteristics of at least one battery set based on the obtained early charge and discharge cycle data, and generating a charge and discharge cycle data feature vector of the battery set to form a corresponding feature vector set; clustering and grouping the charge and discharge cycle data feature vector set using a clustering analysis algorithm; and establishing corresponding regression models for one or more groups of cycle data feature vector subsets after clustering using a regression analysis algorithm, so as to predict the charge and discharge cycle life of the battery according to a regression model among the established regression models that is closest to the characteristics of the battery whose charge and discharge cycle life is to be measured.

[0009] In some embodiments of the first aspect of the present application, the method includes: extracting multiple battery set charge and discharge characteristics based on the acquired early charge and discharge cycle data, which include: any one or more combinations of discharge capacity differential characteristics, discharge curve degradation characteristics, battery internal resistance characteristics, and battery charging cycle number characteristics.

[0010] The charge and discharge cycle life refers to the number of charge and discharge cycles that a battery can perform under a certain capacity. Each charge and discharge is called a charge and discharge cycle. The early charge and discharge cycle data includes a preset number of charge and discharge cycles tested earlier in the total number of charge and discharge cycles of the test battery; the preset number is not fixed and can be determined by the specific test scenario, and this embodiment does not limit this.

[0011] In some embodiments of the first aspect of the present application, a portion of the batteries that have undergone a complete charge and discharge cycle test are used as a training set for establishing a regression model, and another portion of the batteries are used as a test set; the method includes: obtaining multiple early charge and discharge cycle data of the training set; extracting at least one training set charge and discharge feature based on the acquired early charge and discharge cycle data, and generating a charge and discharge cycle data feature vector set of the training set; clustering and grouping the charge and discharge cycle data feature vector set using a clustering analysis algorithm; establishing corresponding regression models for one or more groups of charge and discharge cycle data feature vector subsets after clustering using a regression analysis algorithm; and performing accuracy testing on each regression model based on the test set.

[0012] In some embodiments of the first aspect of the present application, the method includes: acquiring measured data from a battery measurement site; and updating the regression model according to the acquired measured data.

[0013] To achieve the above-mentioned purpose and other related purposes, the second aspect of the present application provides a method for predicting the charge and discharge cycle life of a battery, which includes: obtaining and deploying a regression model for predicting the charge and discharge cycle life of a battery; obtaining and deploying a regression model for predicting the charge and discharge cycle life of a battery; and predicting the charge and discharge cycle life of the battery based on the selected regression model.

[0014] In some embodiments of the second aspect of the present application, the method includes: collecting and uploading actual measured data based on the battery measurement site for updating the regression model; and / or obtaining and deploying one or more updated regression models for predicting the battery charge and discharge cycle life.

[0015] To achieve the above-mentioned purpose and other related purposes, the third aspect of the present application provides a device for predicting the charge and discharge cycle life of a battery, which includes: a data acquisition module for acquiring early charge and discharge cycle data of a battery set; a feature vector set generation module for extracting the charge and discharge characteristics of at least one battery set based on the acquired early charge and discharge cycle data, and generating a feature vector set of the charge and discharge cycle data of the battery set; a clustering module for clustering and grouping the charge and discharge cycle data feature vector set using a clustering analysis algorithm; and a model establishment module for establishing corresponding regression models for one or more groups of charge and discharge cycle data feature vector subsets after clustering using a regression analysis algorithm, so as to predict the charge and discharge cycle life of the battery according to a regression model in the established regression models that is closest to the characteristics of the battery with a charge and discharge cycle life to be measured.

[0016] To achieve the above-mentioned purpose and other related purposes, the fourth aspect of the present application provides a model acquisition and deployment module for acquiring and deploying one or more regression models for predicting the charge and discharge cycle life of a battery; a model selection module for selecting a regression model that is closest to the characteristics of the battery whose charge and discharge cycle life is to be measured from the multiple regression models; and a life prediction module for predicting the charge and discharge cycle life of the battery based on the selected regression model.

[0017] To achieve the above-mentioned purpose and other related purposes, the fifth aspect of the present application provides a computer-readable storage medium, on which a first computer program and / or a second computer program are stored. When the first computer program is executed by a processor, the prediction method provided by the first aspect of the present application is implemented, and when the first computer program is executed by a processor, the prediction method provided by the second aspect of the present application is implemented.

[0018] To achieve the above-mentioned purpose and other related purposes, the sixth aspect of the present application provides an electronic terminal, including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the prediction method provided in the first aspect of the present application.

[0019] To achieve the above-mentioned purpose and other related purposes, the seventh aspect of the present application provides an electronic terminal, including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the prediction method provided in the second aspect of the present application.

[0020] As described above, the battery charge and discharge cycle life prediction method, device, electronic terminal, and storage medium of the present application have the following beneficial effects: the technical solution provided by the present application does not involve the complex physical and chemical mechanism of the battery itself, and only needs to be calculated based on the historical data of the battery to be tested, so it can be easily extended to the prediction of different types of battery life. Compared with other data-driven battery charge and discharge cycle life prediction technologies, the present invention only needs to use the early charge and discharge cycle data of the battery to be tested, and does not need to track all the charge and discharge cycle data of the battery to be tested, so as to realize the prediction of the battery charge and discharge life, which reduces the software and hardware requirements for the battery charge and discharge cycle life prediction system, and is more convenient for deployment and use in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Shown is a flow chart of a system for predicting battery charge and discharge cycle life in one embodiment of the present application.

[0022] Figure 2 Shown is a flow chart of a method for predicting battery charge-discharge cycle life in one embodiment of the present application.

[0023] Figure 3 Shown is a schematic diagram of a regression model set in one embodiment of the present application.

[0024] Figure 4 Shown is a flow chart of a method for predicting battery charge-discharge cycle life in one embodiment of the present application.

[0025] Figure 5 Shown is a flow chart of a method for predicting battery charge-discharge cycle life in one embodiment of the present application.

[0026] Figure 6 Shown is a schematic diagram of the structure of a device for predicting battery charge and discharge cycle life in one embodiment of the present application.

[0027] Figure 7 Shown is a schematic diagram of the structure of a device for predicting battery charge and discharge cycle life in one embodiment of the present application.

[0028] Figure 8 Shown is a schematic diagram of the structure of an electronic terminal in one embodiment of the present application.

[0029] Fig. 9Shown is a schematic diagram of the structure of an electronic terminal in one embodiment of the present application. DETAILED DESCRIPTION

[0030] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0031] As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It should be further understood that the terms "comprise", "include" indicate the presence of the described features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Exceptions to this definition will only occur when the combination of elements, functions, or operations is inherently mutually exclusive in some way.

[0032] In view of the problem that it is difficult to obtain ideal parameters in the existing battery charge and discharge cycle life prediction technology, the present application proposes a prediction method, device, electronic terminal, and storage medium for the battery charge and discharge cycle life. The technical solution provided by the present application does not involve the complex physical and chemical mechanisms of the battery itself, and only needs to be calculated based on the historical data of the battery to be tested. Therefore, it will be relatively easy to generalize to the prediction of different types of battery life. Compared with other data-driven battery charge and discharge cycle life prediction technologies, the present invention only needs to use the early charge and discharge cycle data of the battery to be tested, and does not need to track all the charge and discharge cycle data of the battery to be tested, so as to realize the prediction of the battery charge and discharge cycle life, which reduces the software and hardware requirements for the battery charge and discharge cycle life prediction system, and is more convenient for deployment and use in practical applications.

[0033] like Figure 1 , a flow chart showing a prediction system for battery charge and discharge cycle life in one embodiment of the present application is shown. The prediction system includes a cloud device 11 and a vehicle-mounted device 12 that establish communication connections through a base station.

[0034] The cloud device 11 is used to establish a regression model for predicting the battery charge and discharge cycle life, and the vehicle-mounted terminal device 12 is used to obtain and deploy the regression model from the cloud, and use the regression model to predict the charge and discharge cycle life of the current battery to be tested. The vehicle-mounted terminal device 12 is also used to upload the battery measured data collected from the actual measurement site to the cloud device 11, so that the cloud device 11 can update the regression model based on the battery measured data, and then send the updated regression model to the vehicle-mounted terminal device 12, so as to continuously upgrade the regression model and improve the accuracy of the battery charge and discharge cycle life prediction.

[0035] The cloud device 11 can be a computer including components such as a memory, a storage controller, one or more processing units (CPU), a peripheral interface, an RF circuit, an audio circuit, a speaker, a microphone, an input / output (I / O) subsystem, a display screen, other output or control devices, and an external port; the computer includes but is not limited to personal computers such as desktop computers, laptops, tablet computers, smart phones, smart TVs, personal digital assistants (PDA for short), etc.; the cloud device 11 can also be a server, which can be arranged on one or more physical servers according to various factors such as function and load, or can be composed of distributed or centralized server clusters.

[0036] The vehicle-mounted terminal device 12 is, for example, a vehicle control terminal, which is a front-end device of a vehicle monitoring and management system, also called a vehicle dispatch monitoring terminal (TCU terminal); the vehicle-mounted terminal device 12 may also be a vehicle-mounted controller, which may be, for example, an ARM (Advanced RISC Machines), FPGA (Field Programmable Gate Array), SoC (System on Chip), DSP (Digital Signal Processing), or MCU (Microcontroller Unit) and other controllers.

[0037] In this embodiment, considering that the clustering of the feature vector library and the construction of the regression model involved in the model establishment will consume more computing resources, cloud devices are used to perform the regression model construction task to achieve efficient and rapid modeling. In addition, the application of the measured model to predict the battery charge and discharge cycle life and the collection of measured data consumes less resources, so local vehicle-mounted devices are used to perform, thereby reducing the cost of deployment and application.

[0038] However, it should be noted that in other embodiments, the cloud device used to construct the regression model and the vehicle-mounted device used to predict the battery charge and discharge cycle life can also be designed as an integrated whole, for example, both are set in the cloud or both are set in the vehicle, and this application is not limited to this.

[0039] like Figure 2 , a flow chart showing a method for predicting a battery charge-discharge cycle life in an embodiment of the present application is shown. The method for predicting a battery charge-discharge cycle life in this embodiment is applied to a cloud device, and includes steps S201 to S204.

[0040] In step S201, early charge and discharge cycle data of a battery pack is obtained.

[0041] The basis for building a regression model is the battery set's charge and discharge cycle history data, which should include the complete battery charge and discharge cycle life cycle. Among them, the selection of early charge and discharge cycle data intervals will vary from battery to battery, and needs to be screened and set through experiments. Usually, a set of fixed values ​​can be selected, such as the first 10 cycles or the first 50 cycles, and regression analysis is performed separately. The data interval with the highest accuracy is used to extract early charge and discharge cycle characteristics.

[0042] In step S202, the charge and discharge characteristics of at least one battery set are extracted according to the acquired early charge and discharge cycle data, and a charge and discharge cycle data feature vector set of the battery set is generated accordingly.

[0043] In one embodiment, the extraction of battery charge and discharge curve features includes not only the feature extraction process itself, but also the preprocessing process required for feature extraction, including preprocessing, feature dimension reduction, normalization, etc. Extracting multiple battery charge and discharge features based on the acquired early charge and discharge cycle data includes but is not limited to the following feature combinations: discharge capacity difference feature, discharge curve degradation feature, battery internal resistance feature, or battery charge cycle number feature, etc.

[0044] In step S203, cluster analysis algorithm is used to cluster and group the charge and discharge cycle data feature vector set.

[0045] The cluster analysis algorithm refers to the analysis process of grouping a collection of physical or abstract objects into multiple classes composed of similar objects, and its goal is to collect data on a similar basis for classification. Considering that the effect of data clustering is highly related to the distribution of the feature vector set itself, in the case of insufficient understanding of the distribution of feature vector data, this embodiment provides multiple clustering methods for evaluating the effect of feature clustering, and finally selects the method with the best clustering effect to complete feature clustering.

[0046] Specifically, the cluster analysis algorithm includes but is not limited to: partition clustering algorithm, hierarchical clustering algorithm, fuzzy clustering algorithm, or density-based clustering algorithm, etc.; wherein, the partition clustering algorithm is, for example, K-means algorithm, K-center point algorithm, or CLARANS algorithm, etc.; hierarchical clustering algorithm is, for example, DIANA algorithm, BIRCH algorithm, etc.; fuzzy clustering algorithm is, for example, EM algorithm; density-based algorithm is, for example, OPTICS algorithm, DBSCAN algorithm, etc. There are many types of cluster analysis algorithms, so this embodiment will not be described one by one.

[0047] In step S204, a regression analysis algorithm is used to establish corresponding regression models for one or more groups of charge and discharge cycle data feature vector subsets after clustering, so as to predict the charge and discharge cycle life of the battery according to a regression model among the established regression models that is closest to the characteristics of the battery to be tested.

[0048] The regression analysis algorithm is a predictive analysis algorithm used to study the relationship between the dependent variable (target) and the independent variable (predictor). Regression analysis is performed on each group of cyclic data feature vector subsets after clustering, so as to generate a corresponding regression model set and establish the association relationship between the regression model and each feature class.

[0049] like Figure 3 As shown, a schematic diagram of a regression model set in an embodiment of the present application is shown. The charge and discharge cycle data feature vector set is clustered and grouped into N subsets using a cluster analysis algorithm, namely, charge and discharge cycle data feature vector subset 1, charge and discharge cycle data feature vector subset 2...charge and discharge cycle data feature vector subset N. A corresponding regression model is established for each charge and discharge cycle data feature vector subset using a regression analysis algorithm, that is, regression model 1, regression model 2...regression model N are constructed, and each charge and discharge cycle data feature vector subset corresponds to a regression model.

[0050] In one embodiment, considering that the accuracy of the regression analysis algorithm is highly correlated with the characteristics of the distribution of the feature data set itself, different types of regression analysis methods are used to construct a regression model system, and the final regression model is selected through the optimal regression effect.

[0051] Specifically, the regression analysis algorithm includes but is not limited to the following algorithm combinations: Linear Regression linear regression algorithm, Logistic Regression logistic regression algorithm, Polynomial Regression polynomial regression algorithm, Stepwise Regression stepwise regression algorithm, Ridge Regression ridge regression algorithm, Lasso Regression lasso regression algorithm, ElasticNet regression algorithm, etc.

[0052] In one embodiment, the method further includes: obtaining measured data from a battery test site; and updating the regression model according to the obtained measured data. That is, uploading the early charge and discharge cycle measured data of the local battery in the new energy vehicle to the cloud device for updating the regression model.

[0053] like Figure 4 , a flow chart showing a method for predicting the charge-discharge cycle life of a battery in another embodiment of the present application is shown. In this embodiment, a portion of the multiple batteries that have undergone a complete charge-discharge cycle test is used as a training set for establishing a regression model, and another portion of the batteries is used as a test set; the prediction method includes steps S401 to S405.

[0054] In step S401, a plurality of early charge and discharge cycle data of a training set is obtained.

[0055] In step S402, at least one training set charge and discharge feature is extracted based on the acquired early charge and discharge cycle data, thereby generating a charge and discharge cycle data feature vector set of the training set.

[0056] In step S403, cluster analysis algorithm is used to cluster and group the charge and discharge cycle data feature vector set.

[0057] In step S404, a regression analysis algorithm is used to establish corresponding regression models for the clustered one or more groups of charge-discharge cycle data feature vector subsets.

[0058] In step S405, the accuracy of each regression model is tested based on the test set.

[0059] To facilitate understanding by those skilled in the art, the application principle of the training set and the test set is further explained and illustrated in conjunction with the following embodiments.

[0060] In this embodiment, a batch of batteries (for example, batteries with a number less than 100) is selected, and a complete charge and discharge cycle test is performed on the batteries in the batch until the entire life cycle of the battery ends (the discharge capacity of the battery is less than 80% of the nominal capacity). For the batch of batteries, the batch of batteries is divided into a training set and a test set in a random manner according to a preset ratio.

[0061] Specifically, the charge and discharge features are extracted from the training set and the test set respectively, and a feature vector set of the charge and discharge cycle data of the batch of batteries is constructed. Based on the training set and using a regression analysis algorithm, a battery charge and discharge cycle life prediction model is constructed, and the accuracy of the constructed prediction model is evaluated based on the test set.

[0062] like Figure 5 , which is a flow chart showing a method for predicting a battery charge-discharge cycle life in an embodiment of the present application. The method for predicting a battery charge-discharge cycle life in this embodiment is applied to a vehicle-mounted terminal device, and includes steps S501 to S503.

[0063] In step S501, a regression model for predicting battery charge and discharge cycle life is obtained and deployed.

[0064] In step S502, a regression model that is closest to the characteristics of the battery whose charge-discharge cycle life is to be measured is selected from the multiple regression models.

[0065] Specifically, early charge and discharge test data of the battery whose charge and discharge cycle life is to be tested is obtained, and the same features as those used in building the regression model, i.e., its preprocessing method, are used to obtain normalized input features. A regression model that is closest to the features of the battery to be tested is selected from the regression model set, and the charge and discharge cycle life of the battery to be tested is predicted using the regression model.

[0066] In one embodiment, selecting a regression model from the regression model set that is closest to the characteristics of the battery to be tested refers to analyzing the similarity between the characteristic curve of the battery to be tested and the characteristic curves of each regression model in the regression model set, and taking the regression model with the highest similarity as the regression model that is closest to the characteristics of the battery to be tested.

[0067] Specifically, the similarity between two characteristic curves may be evaluated by distance judgment based on various distance measures, similarity analysis using correlation coefficients, etc., wherein the distance measure may be, for example, Euclidean distance, Hausdorff distance, or Fréchet distance, etc.

[0068] In step S503, the charge-discharge cycle life of the battery is predicted according to the selected regression model.

[0069] In one embodiment, the method also includes collecting and uploading actual measured data based on the battery measurement site for updating the regression model, that is, uploading relevant characteristic data to the battery set historical data set so that the regression model can be automatically updated on the cloud device and sent to the vehicle-mounted device.

[0070] In one embodiment, the method further includes obtaining and deploying one or more updated regression models for predicting the battery charge and discharge cycle life, thereby continuously improving the accuracy of the battery charge and discharge cycle life prediction.

[0071] like Figure 6 , which shows a schematic diagram of the structure of a device for predicting battery charge and discharge cycle life in an embodiment of the present application. The device includes a data acquisition module 61, a feature vector set generation module 62, a clustering module 63, and a model building module 64.

[0072] The data acquisition module 61 is used to acquire early charge and discharge cycle data of a battery set; the feature vector set generation module 62 is used to extract at least one battery set charge and discharge feature based on the acquired early charge and discharge cycle data, and generate a charge and discharge cycle data feature vector set of the battery set; the clustering module 63 is used to cluster and group the charge and discharge cycle data feature vector set using a clustering analysis algorithm; the model establishment module 64 is used to establish corresponding regression models for one or more groups of charge and discharge cycle data feature vector subsets after clustering using a regression analysis algorithm, so as to predict the charge and discharge cycle life of the battery according to a regression model in the established regression model that is closest to the characteristics of the battery whose charge and discharge cycle life is to be measured.

[0073] like Figure 7 , which shows a schematic diagram of a device for predicting the charge and discharge cycle life of a battery in another embodiment of the present application. The device comprises: a model acquisition and deployment module 71, a model selection module 72, and a life prediction module 73.

[0074] The model acquisition and deployment module 71 is used to acquire and deploy one or more regression models for predicting the charge and discharge cycle life of the battery; the model selection module 72 is used to select a regression model that is closest to the characteristics of the battery whose charge and discharge cycle life is to be measured from the multiple regression models; the life prediction module 73 is used to predict the charge and discharge cycle life of the battery based on the selected regression model.

[0075] It should be noted that the implementation of the device for predicting the battery charge-discharge cycle life is similar to the implementation of the method for predicting the battery charge-discharge cycle life, so it will not be repeated. It should be understood that the division of the various modules of the above device is only a division of logical functions, and in actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated.

[0076] like Figure 8 As shown, a schematic diagram of the structure of an electronic terminal in an embodiment of the present application is shown. The electronic terminal provided in this example includes a processor 81 and a memory 82, the memory 82 is connected to the processor 81 through a system bus and completes communication with each other, the memory 82 is used to store computer programs, and the processor 81 is used to run the computer program, so that the electronic terminal executes steps S201 to S204 in the above battery cycle life prediction method.

[0077] like Fig. 9 As shown, a structural schematic diagram of another electronic terminal provided in an embodiment of the present application is shown. The electronic terminal provided in this example includes a processor 91 and a memory 92, the memory 92 is connected to the processor 91 through a system bus and completes communication with each other, the memory 92 is used to store computer programs, and the processor 91 is used to run the computer program, so that the electronic terminal executes steps S501 to S503 in the above battery cycle life prediction method.

[0078] The system bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used to realize the communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM), and may also include non-volatile memory, such as at least one disk storage.

[0079] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0080] In one embodiment, the present application also provides a computer-readable storage medium having a first computer program and / or a second computer program stored thereon, characterized in that when the first computer program is executed by a processor, steps S201 to S204 in the above method for predicting the battery cycle life are implemented, and when the second computer program is executed by a processor, steps S501 to S503 in the above method for predicting the battery cycle life are implemented.

[0081] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0082] In summary, the present application provides a prediction method, device, electronic terminal, and storage medium for the charge and discharge cycle life of a battery. The technical solution provided by the present application does not involve the complex physical and chemical mechanism of the battery itself, and only needs to be calculated based on the historical data of the battery to be tested. Therefore, it will be relatively easy to generalize to the prediction of the charge and discharge life of different types of batteries. Compared with other data-driven battery charge and discharge cycle life prediction technologies, the present invention only needs to use the early charge and discharge cycle data of the battery to be tested, and does not need to track all the charge and discharge cycle data of the battery to be tested, so as to realize the prediction of the charge and discharge cycle life of the battery, thereby reducing the software and hardware requirements for the battery charge and discharge cycle life prediction system, and making it easier to deploy and use in practical applications. Therefore, the present application effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0083] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for predicting battery charge and discharge cycle life, It is characterized in that include: Obtain early charge and discharge cycle data of battery sets; Extracting multiple battery charge and discharge characteristics based on the acquired early charge and discharge cycle data, including: any one or more combinations of discharge capacity difference characteristics, discharge curve degradation characteristics, battery internal resistance characteristics, and battery charge cycle number characteristics; wherein the selection of the early charge and discharge cycle data interval selects a set of fixed values ​​according to different batteries, and uses the data interval with the highest accuracy to extract the early charge and discharge cycle characteristics; Extracting charge and discharge cycle characteristics of at least one battery set according to the acquired early charge and discharge cycle data, and generating a charge and discharge cycle data feature vector of the battery set to form a corresponding feature vector set; Clustering and grouping the charge-discharge cycle data feature vector set using a clustering analysis algorithm; Using a regression analysis algorithm to establish corresponding regression models for one or more groups of charge-discharge cycle data feature vector subsets after clustering, so as to predict the charge-discharge cycle life of the battery according to a regression model that is closest to the characteristics of the battery to be measured in the established regression model; wherein, according to the accuracy of the regression analysis algorithm and the characteristics of the distribution of the feature data set itself, different types of regression analysis algorithms are used to construct the regression model, and the regression model is selected according to the optimal regression effect; Acquire measured data from a battery measurement site; and update the regression model according to the acquired measured data.

2. The method according to claim 1, It is characterized in that A portion of the batteries that have undergone a complete charge-discharge cycle test is used as a training set for establishing a regression model, and another portion of the batteries is used as a test set; the method comprises: Obtain multiple early charge and discharge cycle data of the training set; Extracting at least one training set charge and discharge feature according to the acquired early charge and discharge cycle data, and generating a charge and discharge cycle data feature vector set of the training set; Clustering and grouping the charge-discharge cycle data feature vector set using a clustering analysis algorithm; Using a regression analysis algorithm, a corresponding regression model is established for each of the clustered one or more groups of charge-discharge cycle data feature vector subsets; The accuracy of each regression model is tested based on the test set.

3. A method for predicting battery charge and discharge cycle life, It is characterized in that include: Acquire and deploy a regression model for predicting battery charge and discharge cycle life obtained by the battery charge and discharge cycle life prediction method according to claim 1; Selecting a regression model that is closest to the characteristics of the battery whose charge-discharge cycle life is to be measured from multiple regression models; The charge and discharge cycle life of the battery is predicted based on the selected regression model.

4. The method according to claim 3, It is characterized in that include: Collecting and uploading actual measured data based on the battery actual measurement site for updating the regression model; and / or Obtain and deploy one or more updated regression models for predicting battery charge and discharge cycle life.

5. A device for predicting battery charge and discharge cycle life, It is characterized in that include: A data acquisition module, used to acquire early charge and discharge cycle data of the battery set; Extracting multiple battery charge and discharge characteristics based on the acquired early charge and discharge cycle data, including: any one or more combinations of discharge capacity difference characteristics, discharge curve degradation characteristics, battery internal resistance characteristics, and battery charge cycle number characteristics; wherein the selection of the early charge and discharge cycle data interval selects a set of fixed values ​​according to different batteries, and uses the data interval with the highest accuracy to extract the early charge and discharge cycle characteristics; A feature vector set generation module, used to extract the charge and discharge characteristics of at least one battery set according to the acquired early charge and discharge cycle data, and generate a feature vector set of the charge and discharge cycle data of the battery set; A clustering module, used for clustering and grouping the charge and discharge cycle data feature vector set using a clustering analysis algorithm; The model building module is used to use a regression analysis algorithm to establish corresponding regression models for one or more groups of charge and discharge cycle data feature vector subsets after clustering, so as to predict the charge and discharge cycle life of the battery according to a regression model that is closest to the characteristics of the battery to be measured in the established regression model; obtain measured data from the battery actual measurement site; update the regression model according to the measured data obtained; wherein, according to the accuracy of the regression analysis algorithm and the characteristics of the distribution of the feature data set itself, different types of regression analysis algorithms are used to build the regression model, and the regression model is selected according to the optimal regression effect.

6. A device for predicting battery charge and discharge cycle life, It is characterized in that include: A model acquisition and deployment module, used to acquire and deploy one or more regression models for predicting battery charge and discharge cycle life obtained according to the battery charge and discharge cycle life prediction method of claim 1; A model selection module, used to select a regression model that is closest to the characteristics of the battery whose charge-discharge cycle life is to be measured from the multiple regression models; The life prediction module is used to predict the charge and discharge cycle life of the battery according to the selected regression model.

7. A computer-readable storage medium having stored thereon a first computer program and / or a second computer program, It is characterized in that When the first computer program is executed by the processor, the method for predicting the battery charge and discharge cycle life described in any one of claims 1 to 2 is implemented; when the second computer program is executed by the processor, the method for predicting the battery charge and discharge cycle life described in claim 3 or 4 is implemented.

8. An electronic terminal, It is characterized in that include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory, so that the terminal executes the method for predicting the battery charge and discharge cycle life according to any one of claims 1 to 2.

9. An electronic terminal, It is characterized in that include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the terminal executes the method for predicting the battery charge and discharge cycle life as described in claim 3 or 4.

10. A battery charge and discharge cycle life prediction system, It is characterized in that The electronic terminal comprises the electronic terminal as claimed in claim 8 and the electronic terminal as claimed in claim 9.

Citation Information

Patent Citations

  • Cycle life indirect prediction method for lithium ion battery provided with uncertain intervals on basis of GPR (general purpose register)

    CN103389471A

  • Predication method of electric vehicle power battery service life

    CN103954913A

  • Prediction method for life of secondary battery based on particle filter and mechanism model

    CN106055775A

  • TBM cutting tool life prediction method based on data driven support vector regression machine

    CN106778010A

  • Power battery system residual available life prediction method based on relevance vector machine and particle filter

    CN107238800A