Power battery capacity prediction method, device and equipment
By categorizing power battery data using clustering algorithms and aging models, the method addresses the inaccuracies in existing lifespan prediction methods, enhancing the precision of battery capacity estimation.
Patent Information
- Application Number
- CN202110711103.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-06-25
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2041-06-25
AI Technical Summary
Existing methods for predicting power battery lifespan, such as experimental and model-based approaches, face challenges in accurately estimating battery life due to non-constant current conditions in vehicles and complex real-world driving scenarios, leading to long test cycles and difficulty in obtaining necessary parameters, resulting in inaccurate and inefficient predictions.
A method involving clustering algorithms to categorize power battery data from actual vehicle conditions, using feature identifiers and aging models to predict capacity, and a device comprising an input module for determining the battery state parameters, and a device comprising a computer program, and a computer program, and a computer-readable storage medium.
The method improves the accuracy of battery capacity prediction by distinguishing different aging types and establishing corresponding aging models, providing more accurate and reliable predictions.
Smart Images
Figure CN115526368B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of battery management, and in particular to a power battery capacity prediction method, a power battery capacity prediction device, a power battery capacity prediction equipment and a corresponding storage medium. Background Art
[0002] Under the urgent need of energy conservation and emission reduction in the whole society, the development of new energy vehicles with environmental protection and energy-saving advantages has become a consensus in the automotive industry. As the main direction of the development of new energy vehicles, electric vehicles are facing many challenges and pressures. The performance of power batteries, as an important component of electric vehicles, will directly affect the development and application prospects of electric vehicles. Lifespan is the main performance indicator of power batteries. Accurate prediction of lifespan can not only understand the degradation state of the battery, provide users with accurate vehicle operating status information, and provide a basis for cost calculation of automobile production and manufacturing, but also prevent the occurrence of failure disasters and protect the safety of users' lives and property.
[0003] The life estimation of power batteries is generally achieved by experimental method and model method: the experimental method often uses several constant current charge and discharge cycles to analyze the power battery life through capacity decay, but when the power battery is actually running in a vehicle, the current is not constant, resulting in inaccurate prediction results. In addition, the standard NEDC operating conditions can be used to simulate actual operating conditions for discharge testing, but the test cycle is too long, and a life test takes one year, which cannot meet the development needs of current engineering applications.
[0004] Model methods are mainly divided into mechanism models and statistical models: mechanism models include electrochemical analysis, impedance method, etc. This type of model method mainly uses physical and chemical properties such as battery SOC and SEI film resistance as input parameters, and predicts battery life by quantitatively studying the attenuation of power battery capacity. Statistical models are mainly life prediction models based on neural network design. For example, patents: CN201910394691.0 and CN202011087935.X. The actual vehicle driving conditions are complex, the mechanism model parameters are difficult to obtain, and it is difficult to accurately predict battery life. Laboratory tests cannot exhaust the aging data of real working conditions. Summary of the invention
[0005] In view of this, the present invention aims to propose a method, device and equipment for predicting the capacity of a power battery, so as to at least partially solve the problems in the prior art of long test cycle of the experimental method, difficulty in obtaining parameters of the model method and high model complexity.
[0006] To achieve the above-mentioned purpose, the present invention provides a method for predicting the capacity of a power battery, the prediction method comprising: obtaining sample data of a power battery; dividing the sample data into several categories through a clustering algorithm, wherein each category has a corresponding aging model and a feature identifier; the aging model is obtained through the following steps: determining a fitting relationship in the aging model; determining parameters in the fitting relationship through sample data of a corresponding type of the aging model; the feature identifier is used to identify the features of the sample data of the corresponding category; obtaining battery state parameters of a power battery to be tested; determining an aging model adopted by the battery state parameters from a plurality of aging models; and inputting the battery state parameters into the adopted aging model to obtain the corresponding battery capacity.
[0007] Preferably, the sample data of the power battery includes: multiple groups of historical data of the same type of power battery in actual vehicle driving conditions.
[0008] Preferably, the sample data is divided into several categories by a clustering algorithm, including: presetting a clustering algorithm and determining clustering parameters in the clustering algorithm; dividing the sample data into core points or boundary points according to the clustering parameters; constructing categories according to the core points, and dividing the sample data into the several categories.
[0009] Preferably, the clustering algorithm is a DBSCAN algorithm; the clustering parameters include a neighborhood radius and a neighborhood point count threshold.
[0010] Preferably, the feature identifier is a cluster center; the feature identifier is a cluster center; determining the aging model adopted by the battery status parameter from multiple aging models, including: calculating the distance between the battery status parameter and the cluster center corresponding to each category; selecting the aging model with the closest distance as the aging model adopted by the battery status parameter.
[0011] Preferably, the fitting relationship includes polynomial fitting, neural network fitting or regression tree fitting.
[0012] Preferably, the battery status parameters include at least two of: current, voltage, temperature, state of charge, storage time, depth of discharge, and Cooley efficiency.
[0013] In the second aspect of the present invention, a power battery capacity prediction device is also provided, including: an input module for obtaining battery status parameters; a matching module for matching with the battery status parameters to determine an aging model adopted by the battery status parameters from multiple aging models; the multiple aging models correspond one-to-one to several categories into which sample data are divided by a clustering algorithm; each of the aging models includes a mapping relationship between battery status parameters and battery capacity; and a calculation module for inputting the battery status parameters into the adopted aging model to obtain the corresponding battery capacity.
[0014] In the third aspect of the present invention, a power battery capacity prediction device is also provided, comprising: at least one processor; a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the at least one processor implements the aforementioned power battery capacity prediction method by executing the instructions stored in the memory.
[0015] In a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the aforementioned power battery capacity prediction method is implemented.
[0016] Compared with the prior art, the power battery capacity prediction method, device and equipment provided by the embodiments of the present invention have the following beneficial effects:
[0017] The implementation method provided by the present invention can distinguish multiple types of aging, and then establish corresponding aging models for different aging types, thereby improving the accuracy of the aging model and achieving more accurate prediction of the power battery capacity. When the sample data is richer, the more data of different aging types, the higher the coverage of different aging types, the more comprehensive the clustering algorithm is in distinguishing different types, and the more intuitive and reliable the distinction effect is.
[0018] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:
[0020] Figure 1 is a schematic flow chart of a power battery capacity prediction method provided in an embodiment of the present invention;
[0021] Figure 2 is a flow chart of a clustering algorithm in a power battery capacity prediction method provided in an embodiment of the present invention;
[0022] Figure 3 is a schematic diagram of calculating the cluster center distance in the power battery capacity prediction method provided in an embodiment of the present invention;
[0023] Figure 4 It is a schematic diagram of the implementation process of the power battery capacity prediction method provided by the embodiment of the present invention;
[0024] Figure 5 It is a schematic diagram of the structure of a power battery capacity prediction device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0025] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0026] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.
[0027] Figure 1 is a flow chart of a method for predicting the capacity of a power battery provided by an embodiment of the present invention, such as Figure 1 A method for predicting the capacity of a power battery, the method comprising:
[0028] S01. Obtain sample data of the power battery;
[0029] The sample data here include sample data of the power battery under actual vehicle driving conditions, including current I, voltage V, temperature T, state of charge SOC, storage time t, depth of discharge DOD, Curry efficiency μ, etc., or select parameters therein for combination.
[0030] S02, dividing the sample data into several categories by a clustering algorithm, and determining a corresponding aging model and feature identifier for each category; the aging model is obtained by the following steps: determining a fitting relationship in the aging model; determining parameters in the fitting relationship by using sample data of a type corresponding to the aging model; the feature identifier is used to identify features of the sample data of a corresponding category;
[0031] The sample data is classified, and each category of sample data has certain similarities or internal connections. By classifying the sample data through a clustering algorithm, the classification results can be obtained quickly and better classification results can be achieved. The clustering algorithm here can be selected from the existing clustering algorithms according to actual needs. An aging model needs to be determined for each category. The aging model is a mathematical model. First, it is necessary to determine the fitting relationship in the mathematical model, that is, to select a suitable fitting function. Then, the fitting relationship is trained or corrected through the sample data belonging to this type to determine the parameters in the fitting relationship, thereby obtaining an aging model that calculates the battery capacity by inputting parameters.
[0032] S03, obtaining battery status parameters of the power battery to be tested;
[0033] The battery state parameters obtained here are used as input parameters for predicting capacity, and the parameters included are the same as or have an inclusion relationship with the parameters in the sample data in step S01.
[0034] S04, determining an aging model adopted by the battery state parameter from a plurality of aging models;
[0035] For specific battery status parameters, the aging model used needs to be determined first. The same battery status parameter can obtain different battery capacities according to different aging models. This embodiment determines the aging model used by feature identification. By determining the appropriate aging model, a more accurate battery capacity can be calculated.
[0036] S05, input the battery state parameters into the adopted aging model to obtain the corresponding battery capacity. The aging model adopted in this step is the aging model determined in step S04, which is used to obtain the corresponding battery capacity according to the battery state parameters, that is, inputting the battery state parameters can obtain the corresponding battery capacity.
[0037] Through the above implementation, multiple types of aging types can be distinguished, and then corresponding aging models can be established for different aging types, thereby improving the accuracy of the aging model and achieving more accurate prediction of the power battery capacity. When the sample data is richer, the more data of different aging types, the higher the coverage of different aging types, the more comprehensive the clustering algorithm is in distinguishing different types, and the more intuitive and reliable the distinction effect is.
[0038] In one embodiment of the present invention, the sample data of the power battery includes: multiple sets of historical data of the same type of power battery in actual vehicle driving conditions. Using historical data in actual vehicle driving conditions as samples can better fit the real scene, and is also conducive to the acquisition of massive samples. By using a large amount of sample data reflecting the actual status of several power batteries, the defects of the experimental method and the model method can be overcome, clustering can be made more accurate, and thus the battery capacity prediction can be made more accurate.
[0039] Figure 2 is a flow chart of a clustering algorithm in a power battery capacity prediction method provided in an embodiment of the present invention, such as Figure 2As shown. In this embodiment, the sample data is divided into several categories by a clustering algorithm, including: presetting a clustering algorithm and determining clustering parameters in the clustering algorithm; dividing the sample data into core points or boundary points according to the clustering parameters; constructing categories according to the core points, and dividing the sample data into the several categories. Further, the clustering algorithm is a DBSCAN algorithm; the clustering parameters include a neighborhood radius Eps and a neighborhood point number threshold Minpts. The specific process is as follows: In one embodiment, first scan the data set, select the unvisited point p, generate the neighborhood set Np, if the number of points in Eps(p) is greater than Minpts, then p is determined to be a core point, and a new cluster C is generated, then select the unclassified point q in Np, if q has not been visited, generate the neighborhood set Nq, if the number of points in Eps(q) is greater than Minpts, then q is determined to be a core point, update Np=Np+Nq, point q is added to cluster C, if q is not a core point, and q has no category, then q is determined to be a boundary point, added to cluster C, and so on until Np no longer contains unclassified points, then if the data set D still contains unvisited points, return to the second step, select unvisited points. Through the above method, the sample data is divided into the aforementioned several categories.
[0040] Figure 3 is a schematic diagram of calculating the cluster center distance in the power battery capacity prediction method provided in an embodiment of the present invention, such as Figure 3 As shown. In this embodiment, the feature identifier is a cluster center; determining the aging model used for the battery state parameter from multiple aging models includes: calculating the distance between the battery state parameter and the cluster center corresponding to each category; selecting the aging model with the closest distance as the aging model used for the battery state parameter. The figure only shows the case of four cluster centers C1-C4, and their number does not constitute a limitation on the number of types. Common clustering calculations in cluster analysis include Euclidean distance, Manhattan distance, Chebyshev distance, etc., which are mainly used for similarity measurement. Through the calculation of the above method, the similarity (distance) between two objects can be obtained. In actual calculations, effective selection should be made according to the attribute characteristics of different objects. The distance calculation of objects is very important for the process of clustering algorithm and directly affects the effectiveness of the algorithm, so it should be carefully selected in actual selection.
[0041] In one embodiment of the present invention, the fitting relationship includes polynomial fitting, neural network fitting or regression tree fitting. The polynomial fitting includes: y = p_{0}x^n+p_{1}x^{n-1}+p_{2}x^{n-2}+p_{3}x^{n-3}+...+p_{n}; the number of terms of the polynomial can be determined as needed. Neural network fitting includes: CNN, RNN and GAN, etc., select a suitable neural network structure, and train it with sample data to obtain an aging model that can predict the battery capacity. Regression tree fitting includes commonly used binary trees, for example: using a binary tree to recursively divide the prediction space into several subsets, and the distribution of Y in these subsets is continuous and uniform. The leaf nodes in the tree correspond to different divided areas, and the division is determined by the branching rules (Spitting Rules) associated with each internal node. By moving from the root to the leaf node, a prediction sample is assigned to a unique leaf node, and the conditional distribution of Y at the node is also determined. The specific steps for establishing different aging models are not repeated here.
[0042] In one embodiment of the present invention, the battery status parameters include at least two of current, voltage, temperature, state of charge, storage time, depth of discharge, and curry efficiency. The more parameters are input, the more accurate the battery capacity is. In a specific scenario, a person skilled in the art selects at least two of the above-mentioned battery status parameters to combine according to actual conditions and measurement conditions to obtain a more accurate battery capacity.
[0043] Figure 4 is a schematic diagram of the implementation process of the power battery capacity prediction method provided by the embodiment of the present invention, such as Figure 4 In this embodiment, the power battery capacity prediction method includes the following steps:
[0044] (1) Preset clustering algorithm, parameters and related thresholds.
[0045] (2) Input sample data of the power battery under actual vehicle driving conditions.
[0046] (3) The sample data is divided into several categories through the clustering algorithm, and the cluster centers C1, ..., Ck are obtained.
[0047] (4) The sample data of each category is used to obtain the model parameters of the corresponding aging model through statistical models such as polynomial fitting, neural network, and regression tree.
[0048] Cap=f(I,V,T,SOC,t,DOD,μ)
[0049] (5) Compare the distances between the test data and each cluster center to determine which aging model it belongs to.
[0050] (6) Input the test data into the corresponding aging model to calculate the power battery capacity.
[0051] Figure 5 is a schematic diagram of the structure of a power battery capacity prediction device provided by an embodiment of the present invention, such as Figure 5 In this embodiment, a power battery capacity prediction device includes: an input module for acquiring battery state parameters; a matching module for matching with the battery state parameters to determine an aging model adopted by the battery state parameters from a plurality of aging models; the plurality of aging models correspond one to one to a plurality of categories into which sample data is divided by a clustering algorithm; each of the aging models includes a mapping relationship between a battery state parameter and a battery capacity; and a calculation module for inputting the battery state parameters into the adopted aging model to obtain the corresponding battery capacity.
[0052] The specific definition of each functional module in the above-mentioned power battery capacity prediction device can refer to the definition of the power battery capacity prediction method above, which will not be repeated here. Each module in the above-mentioned device can be implemented in whole or in part by software, hardware and a combination thereof. 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 operations corresponding to the above modules.
[0053] In one embodiment of the present invention, a power battery capacity prediction device is also provided, comprising: at least one processor; a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the at least one processor implements the aforementioned power battery capacity prediction method by executing the instructions stored in the memory. The control module or processor here has the functions of numerical calculation and logical operation, and it has at least a central processing unit CPU with data processing capability, a random access memory RAM, a read-only memory ROM, a variety of I / O ports and an interrupt system, etc. The processor includes a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to one or more, and the aforementioned method is implemented by adjusting the kernel parameters. The memory may include a non-permanent memory in a computer-readable medium, a random access memory (RAM) and / or a non-volatile memory, such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory includes at least one storage chip.
[0054] In one embodiment of the present invention, a computer-readable storage medium stores a computer program, which implements the aforementioned power battery capacity prediction method when executed by a processor.
[0055] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0056] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0057] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0058] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0059] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0060] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0061] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0062] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0063] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A method for predicting power battery capacity, characterized in that: The prediction method comprises: Obtain sample data of power batteries; The sample data is divided into several categories by a clustering algorithm, wherein each category has a corresponding aging model and feature identifier; the aging model is obtained by the following steps: determining a fitting relationship in the aging model; determining parameters in the fitting relationship by sample data of a type corresponding to the aging model; the feature identifier is used to identify features of sample data of a corresponding category; the feature identifier is a cluster center; wherein the sample data is divided into several categories by a clustering algorithm, including: presetting a clustering algorithm and determining clustering parameters in the clustering algorithm; dividing the sample data into core points or boundary points according to the clustering parameters; constructing categories according to the core points, and dividing the sample data into the several categories, the clustering algorithm is a DBSCAN algorithm; the clustering parameters include a neighborhood radius and a neighborhood point number threshold; Obtaining battery status parameters of the power battery to be tested; Determining the aging model adopted by the battery state parameter from a plurality of aging models includes: calculating the distance between the battery state parameter and the cluster center corresponding to each category; selecting the aging model with the closest distance as the aging model adopted by the battery state parameter; The battery state parameters are input into the adopted aging model to obtain the corresponding battery capacity.
2. The prediction method according to claim 1, characterized in that: The sample data of the power battery includes: multiple groups of historical data of the same type of power battery in actual vehicle driving conditions.
3. The prediction method according to claim 1, characterized in that: The fitting relationship includes polynomial fitting, neural network fitting or regression tree fitting.
4. The prediction method according to claim 1, characterized in that: The battery status parameters include: at least two of: current, voltage, temperature, state of charge, storage time, depth of discharge, and Cooley efficiency.
5. A power battery capacity prediction device, characterized in that: include: Input module, used to obtain battery status parameters; A matching module, used for matching with the battery state parameter to determine the aging model adopted by the battery state parameter from a plurality of aging models; the plurality of aging models correspond one to one with a plurality of categories into which the sample data is divided by a clustering algorithm; each of the aging models includes a mapping relationship between the battery state parameter and the battery capacity, and uses the cluster center as a characteristic identifier; Determining the aging model used by the battery state parameter from multiple aging models, including: calculating the distance between the battery state parameter and the cluster center corresponding to each category; selecting the aging model with the closest distance as the aging model used by the battery state parameter; wherein the sample data is divided into several categories by a clustering algorithm, including: presetting a clustering algorithm and determining clustering parameters in the clustering algorithm; dividing the sample data into core points or boundary points according to the clustering parameters; constructing categories according to the core points, and dividing the sample data into the several categories, the clustering algorithm is a DBSCAN algorithm; the clustering parameters include a neighborhood radius and a neighborhood point number threshold; and The calculation module is used to input the battery state parameters into the adopted aging model to obtain the corresponding battery capacity.
6. A power battery capacity prediction device, characterized in that: include: at least one processor; a memory connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the at least one processor implements the power battery capacity prediction method according to any one of claims 1 to 4 by executing the instructions stored in the memory.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the power battery capacity prediction method described in any one of claims 1 to 4 is implemented.