Battery cell health degree calculation method, control device and storage medium

Abstract

The invention relates to a cell health degree calculation method, a control device and a storage medium, and the method comprises the steps: carrying out the calculation of a cell through employing a cell health degree algorithm with the accuracy greater than or equal to a threshold value, and obtaining the cell health degree; the process of obtaining the accuracy of the cell health degree algorithm comprises the following steps: obtaining a first cell sample set, wherein cells in the first cell sample set are similar samples; based on the first battery cell sample set, first probability distribution of the battery cell health degree corresponding to the first battery cell sample set is obtained, and the first probability distribution is probability distribution obtained based on the real value of the battery cell health degree; based on the first battery cell sample set and a battery cell health degree algorithm, second probability distribution of the battery cell health degree corresponding to the first battery cell sample set is obtained, and the second probability distribution is probability distribution obtained based on the battery cell health degree algorithm; and obtaining the accuracy of the cell health degree algorithm based on the first probability distribution and the second probability distribution.

Description

Technical Field

[0001] The present invention relates to the field of battery health, and specifically provides a method for calculating the health of a battery cell, a control device, and a storage medium. Background Art

[0002] With the popularization of electric vehicles, the health of the batteries in electric vehicles affects the safety of the vehicle's cruising range. Therefore, the accuracy of calculating the battery health has become a problem to be solved.

[0003] In the prior art, the battery health is calculated through algorithms. However, the battery health values given by different algorithms are different, and the deviations of the battery health values given by these algorithms vary. In the prior art, only an algorithm considered appropriate can be selected through experience to obtain the battery health, but the battery health obtained in this way cannot be guaranteed to be the closest to the true value. Then how to obtain a relatively accurate battery health has become an urgent problem to be solved in this field.

[0004] Correspondingly, a new solution for calculating the health of a battery cell is needed in this field to solve the above problems. Summary of the Invention

[0005] In order to overcome the above defects, the present invention is proposed to provide a solution to solve or at least partially solve the technical problem in the prior art that it is not easy to determine which battery cell health algorithm is more suitable.

[0006] In a first aspect, the present invention provides a method for calculating the health of a battery cell, including: obtaining a first battery cell sample set, where the battery cells in the first battery cell sample set are all similar samples to each other; based on the first battery cell sample set, obtaining a first probability distribution of the battery cell health corresponding to the first battery cell sample set, where the first probability distribution is a probability distribution based on the true value of the battery cell health; based on the first battery cell sample set and a battery cell health algorithm, obtaining a second probability distribution of the battery cell health corresponding to the first battery cell sample set, where the second probability distribution is a probability distribution based on the battery cell health algorithm; based on the first probability distribution and the second probability distribution, obtaining the accuracy of the battery cell health algorithm.

[0007] In the above technical solution, "based on the first battery cell sample set, obtaining a first probability distribution of the battery cell health corresponding to the first battery cell sample set" includes: based on the first battery cell sample set, obtaining a second battery cell sample set, where the second battery cell sample set is a subset of the first battery cell sample set; based on the second battery cell sample set, obtaining a data set of the battery cells in the second battery cell sample set, where the data set includes a first battery cell health, and the first battery cell health is the true value of the battery cell health; based on the data set, obtaining the first probability distribution.

[0008] In one of the above technical solutions, "obtaining the first probability distribution based on the data set" includes: obtaining a true state-of-health model of the second cell sample set based on the data set, where the true state-of-health model of the battery reflects the relationship between the state of health of the cell and the variables affecting the state of health, and the data set includes the values of the variables affecting the state of health corresponding to the first state of health of the cell; obtaining the values of the variables affecting the state of health corresponding to the cells in the first cell sample set; obtaining the second state of health of the cells in the first cell sample set based on the true state-of-health model of the battery and the values of the variables affecting the state of health; and obtaining the first probability distribution based on the second state of health of the cells.

[0009] In one of the above technical solutions, "obtaining the second probability distribution corresponding to the state of health of the cells in the first cell sample set based on the first cell sample set and the cell state-of-health algorithm" includes: calculating the cells in the first cell sample set based on the cell state-of-health algorithm to obtain the third state of health of the cells in the first cell sample set; and obtaining the second probability distribution based on the third state of health of the cells.

[0010] In one of the above technical solutions, both the first probability distribution and the second probability distribution include the probability distribution of the state of health of the cell corresponding to the same time node of the cell.

[0011] In one of the above technical solutions, both the first probability distribution and the second probability distribution include the probability distributions of the cell corresponding to at least two time nodes.

[0012] In one of the above technical solutions, "obtaining the accuracy of the cell state-of-health algorithm based on the first probability distribution and the second probability distribution" includes: determining the distribution type of the first probability distribution, and obtaining an accuracy calculation method based on the distribution type; and obtaining the accuracy based on the first probability distribution, the second probability distribution, and the accuracy calculation method.

[0013] In one of the above technical solutions, "obtaining the accuracy of the cell state-of-health algorithm based on the first probability distribution and the second probability distribution" includes: for each time node, obtaining the similarity corresponding to each time node based on the probability distribution in the first probability distribution corresponding to the same time node and the probability distribution in the second probability distribution corresponding to the same time node; and obtaining the accuracy of the cell state-of-health algorithm based on the similarity corresponding to each time node.

[0014] In one of the above technical solutions, after "obtaining the accuracy of the battery cell health algorithm", the method further includes: if the accuracies corresponding to multiple battery cell health algorithms are obtained, then based on the accuracy and a preset condition, a battery cell health algorithm after being screened by the preset condition is obtained.

[0015] In one of the above technical solutions, the health degree influence variables at least include one or more parameters among working temperature, discharge rate, discharge depth, and fast charge rate.

[0016] In a second aspect, the present invention provides a device for evaluating the advantages and disadvantages of different battery cell health algorithms, including:

[0017] A battery cell sample sampling module: obtaining a first battery cell sample set, where the battery cells in the first battery cell sample set are all similar samples to each other;

[0018] A true probability distribution module: based on the first battery cell sample set, obtaining a first probability distribution of the battery cell health corresponding to the first battery cell sample set, where the first probability distribution is a probability distribution obtained based on the true value of the battery cell health;

[0019] An algorithm probability distribution module: based on the first battery cell sample set and a battery cell health algorithm, obtaining a second probability distribution of the battery cell health corresponding to the first battery cell sample set, where the second probability distribution is a probability distribution obtained based on the battery cell health algorithm;

[0020] An analysis module: based on the first probability distribution and the second probability distribution, obtaining the accuracy of the battery cell health algorithm.

[0021] In a third aspect, a control device is provided. The control device includes a processor and a storage device. The storage device is adapted to store multiple computer programs, and the computer programs are adapted to be loaded and run by the processor to execute the battery cell health calculation method described in any one of the technical solutions of the above battery cell health calculation method technical solutions.

[0022] In a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores multiple computer programs, and the computer programs are adapted to be loaded and run by a processor to execute the battery cell health calculation method described in any one of the technical solutions of the above battery cell health calculation method technical solutions.

[0023] One or more of the above technical solutions of the present invention have at least one or more of the following Beneficial effects:

[0024] In implementing the technical solution of the present invention, by selecting a cell health algorithm corresponding to an accuracy greater than or equal to a threshold, the accuracy of the electric vehicle battery health is significantly improved.

[0025] Furthermore, first, by obtaining a first set of cell samples with high consistency and comparing the first probability distribution constructed based on actual measurement values with the second probability distribution predicted based on different algorithms, the accuracy of the algorithm is objectively evaluated. This process makes the calculation of battery health no longer rely on empirical judgment, but on data-driven and scientific analysis, ensuring the accuracy and credibility of the calculation results. This technology effectively improves the quality of performance evaluation of electric vehicle batteries, provides strong data support for the safe operation and maintenance of electric vehicles, and helps to improve the overall reliability of electric vehicles and the driving experience of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Referring to the accompanying drawings, the disclosure of the present invention will become more understandable. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the protection scope of the present invention. In addition, similar numbers in the figures are used to represent similar components, where:

[0027] Figure 1 is a schematic flowchart of the main steps of a method for calculating cell health according to an embodiment of the present invention;

[0028] Figure 2 is a schematic flowchart of the secondary steps of a method for calculating cell health according to an embodiment of the present invention;

[0029] Figure 3 is a schematic flowchart of the secondary steps of a method for calculating cell health according to an embodiment of the present invention;

[0030] Figure 4 is a schematic flowchart of the steps of an implementation manner of a method for calculating cell health according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The following describes some embodiments of the present invention with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principle of the present invention and are not intended to limit the protection scope of the present invention.

[0032] In the description of the present invention, a "module" and a "processor" may include hardware, software, or a combination of both. A module may include a hardware circuit, various suitable sensors, communication ports, a memory, and may also include a software part, such as a computer program, or may be a combination of software and hardware. A processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, in hardware, or in a combination of both. A non-transitory computer-readable storage medium includes any suitable medium for storing a computer program, such as a magnetic disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, and so on.

[0033] In the present embodiment, the following method is used to compare the cell health more accurately, specifically as follows:

[0034] Calculate the cell health of the cell using the cell health algorithm whose accuracy is greater than or equal to the threshold.

[0035] In the present embodiment, the threshold here can be either a determined preset threshold or a threshold customized according to the requirements of scientific research personnel.

[0036] In the field of battery management technology, there are various cell health algorithms for predicting and evaluating the current state and future performance of cells. However, determining which algorithm is most suitable for a specific cell poses a challenge.

[0037] In the present embodiment, select a cell health algorithm whose accuracy is greater than or equal to the threshold to calculate the cell, so as to select a satisfactory algorithm from various cell health algorithms, and thus obtain a more accurate cell health.

[0038] In the present embodiment, accuracy is a parameter for judging the difference between the cell health obtained by the cell health algorithm and the true value of the cell health. The higher the accuracy, the smaller the difference between the two. Naturally, selecting a cell health algorithm with an accuracy greater than or equal to the threshold will result in a cell health that meets the requirements.

[0039] Refer to the appendix Figure 1 , Figure 1 is a schematic diagram of the main step flow of the calculation method of cell health according to an embodiment of the present invention. As Figure 1 shown, the calculation method of cell health in the embodiment of the present invention mainly includes the following steps S10-step S40.

[0040] Step S10: Obtain the first cell sample set.

[0041] In the present embodiment, the cells in the first cell sample set are all similar samples to each other.

[0042] In one embodiment, obtaining the first battery cell sample set is to increase the accuracy and reliability of subsequent evaluations. In this embodiment, each battery cell in the first battery cell sample set is a similar sample.

[0043] In this embodiment, a clustering method based on the battery cell production off-line data is adopted to ensure the representativeness and consistency of the sample set. Clustering is a statistical classification technique aimed at grouping objects with similar characteristics into the same group. Preferably, in one embodiment, the production batch, capacity, internal resistance, and self-discharge rate parameters of the battery cell are used as the basis for clustering.

[0044] The reason for choosing these parameters is based on their high importance in battery performance and life evaluation. Specifically as follows:

[0045] Production batch: Provides information about the manufacturing date and conditions of the battery cell, which helps to identify potential differences caused by the production process.

[0046] Capacity: Reflects the ability of the battery cell to store electrical energy, and the aging of the battery usually affects the capacity of the battery cell.

[0047] Internal resistance: An inherent characteristic of the battery, and the aging of the battery cell usually affects the change of the internal resistance.

[0048] Self-discharge rate: The aging of the battery cell usually affects the self-discharge rate, and an increase in the self-discharge rate is an omen of a decline in battery performance.

[0049] Through cluster analysis, a group of similar samples is identified from a large number of battery cells, which constitutes the first battery cell sample set. In this embodiment, the battery cells in the first battery cell sample set are all similar, that is, all the data of the battery cells in the first battery cell sample set are extremely similar. In other words, this batch of samples has a relatively high similarity. To make it easier to understand this basis, a simple example is given here. Suppose there are two battery cells, and these two battery cells are not similar, then it will cause different data during their use. For example, after releasing the same amount of electricity, the remaining power of these two battery cells is different. After using for a period of time under the same scenario and the same conditions, the battery health of these two battery cells may have a large difference due to different constitutions. For example, one battery cell may maintain a high efficiency due to a low internal resistance, while the other battery cell may have a high internal resistance due to minor differences during manufacturing, resulting in a low efficiency. Therefore, after a long time of use, the health of the latter may decline faster than that of the former.

[0050] In contrast, in the first set of cell samples, since all cells are selected and have highly similar characteristics, such as similar capacity, internal resistance, self-discharge rate, etc., they exhibit similar performance and attenuation characteristics under the same usage conditions. This consistency makes the data obtained from these cells more reliable and more representative for the evaluation of the cell health algorithm.

[0051] It should be noted that in the field of electric vehicles, a battery pack consists of multiple cells, and each cell is an essential basic unit in the battery pack. The performance of the cells directly affects the performance of the entire battery pack. Therefore, selecting cells with strong representativeness and consistent performance is crucial for the entire evaluation process.

[0052] Step S20: Based on the first set of cell samples, obtain the first probability distribution corresponding to the cell health of the first set of cell samples.

[0053] In this embodiment, the first probability distribution is a probability distribution obtained based on the true value of the cell health.

[0054] In one implementation, the establishment of the first probability distribution is based on the actual measured values of the cell health, and these values reflect the true performance and state of the cells during their life cycle. In this implementation, the first probability distribution is a probability distribution constituted by all or part of the cells in the first set of cell samples as the data basis, that is, the first probability distribution is a probability distribution constituted by the cell health of all or part of the cells in the first set of cell samples.

[0055] In the field of battery technology, the health of cells is usually affected by various factors, including but not limited to the number of charge-discharge cycles, operating temperature, depth of discharge, etc. The health of each cell changes over time and usage conditions. Therefore, accurately measuring and recording these changes is crucial for understanding the long-term performance of the battery.

[0056] By establishing the first probability distribution based on the true value, step S20 provides a solid benchmark for the subsequent algorithm evaluation. The first probability distribution not only reflects the health status of individual cells but also reveals the trends and distributions of the cell health of all or part of the cells in the entire sample set. Compared with a single data point, the probability distribution can more comprehensively describe the change range and general characteristics of the cell health, thus minimizing misunderstandings and errors caused by the special circumstances of individual cells.

[0057] In addition, the existence of the first probability distribution makes the evaluation of the cell health algorithm in the subsequent steps more objective and effective. By comparing the probability distribution of the cell health generated by the algorithm with the first probability distribution, the performance of the algorithm can be accurately evaluated, and the accuracy and reliability of the algorithm in predicting the cell health can be identified.

[0058] In summary, step S20 lays a foundation for the accuracy evaluation of the battery cell health algorithm by establishing a probability distribution of the battery cell health based on real measurement values, that is, the first probability distribution.

[0059] In this embodiment, a specific first probability distribution is obtained through steps S201 - S203, as Figure 2 shown below:

[0060] Step S201: Obtain a second battery cell sample set based on the first battery cell sample set.

[0061] In this embodiment, the second battery cell sample set is a subset of the first battery cell sample set

[0062] In one embodiment, the purpose of this step is to evaluate the health of the battery cell under more controlled and precise experimental conditions. To ensure the feasibility and effectiveness of the experiment, the scale of the second battery cell sample set is smaller than that of the first battery cell sample set.

[0063] In this embodiment, a specific scenario is given, where it is assumed that the scale of the first battery cell sample set is between 100n and 200n, where n is not less than 30. This means that the first battery cell sample set may contain from 3000 to 6000 battery cells. From this large batch of samples, at least 30 battery cells are randomly selected to form the second battery cell sample set for subsequent detailed experimental analysis. This selection process uses random selection to ensure the randomness of the samples, so as to ensure the representativeness and consistency of the selected samples.

[0064] The reason for selecting a smaller number of battery cells as the second battery cell sample set is that tests under laboratory conditions usually involve more resources and time. Each battery cell needs to undergo a series of tests to accurately determine its health. These tests may include, but are not limited to, charge and discharge cycles, battery cell capacity measurement, and internal resistance test. Due to the complexity of the experiment and cost considerations, it is impossible to conduct the same detailed experiments on all battery cells in the first battery cell sample set.

[0065] In addition, by conducting detailed tests on this subset of battery cells in the laboratory environment, more in - depth and accurate data on the actual performance of the battery cells can be obtained. These data provide a solid foundation for constructing the first probability distribution, ensuring the accuracy and reliability of subsequent analysis.

[0066] Generally speaking, step S201 provides the necessary data and information for obtaining the true value of the battery cell health by screening out a small - scale second battery cell sample set from the larger first battery cell sample set and conducting detailed tests under controlled laboratory conditions

[0067] Step S202: Based on the second battery cell sample set, obtain the data set of the battery cells in the second battery cell sample set, where the data set includes the first battery cell health degree.

[0068] In this embodiment, the first battery cell health degree is the true value of the battery cell health degree.

[0069] In one implementation manner, through an experimental method, key data is obtained based on the battery cells in the second battery cell sample set, and thus a data set including the true value of the battery cell health degree, that is, a data set including the first battery cell health degree, is constructed.

[0070] In one implementation manner, the data set only includes the first battery cell health degree. In this case, the extracted data set directly reflects the actual health state of the battery cells. The first battery cell health degrees corresponding to these battery cells are obtained based on a series of laboratory tests, such as charge and discharge cycles, voltage and current measurements, etc. Through these data, the first probability distribution corresponding to the first battery cell sample set can be directly constructed, and this probability distribution will directly represent the actual change and distribution range of the battery cell health degree.

[0071] It should be noted that the data in the data set described here does not include the time data corresponding to the first battery cell health degree. If the data set not only includes the first battery cell health degree, and each first battery cell health degree has its own corresponding time data, then in the present invention, it is considered to be the case where the data set only includes the first battery cell health degree.

[0072] In another implementation manner, in addition to including the first battery cell health degree, the data set also includes other relevant data of the battery cells corresponding to the battery cells when collecting the first battery cell health degree and other factors affecting the battery cell life. These additional data include some working performances of the battery cells corresponding to the first battery cell health degree. For example, the discharge rate or fast charge rate of the battery cell corresponding to a certain first battery cell health degree when collected. The existence of these data enables a deeper understanding of the relationship between the battery cell health degree and these variables through modeling. For example, statistical or machine learning methods can be used to explore how these variables jointly affect the health status of the battery cells. Since the second battery cell sample set is selected from the first battery cell sample set, these models are not only applicable to the second sample set, but also can be extended and applied to a larger-scale first sample set, thereby providing a more comprehensive first probability distribution.

[0073] Step S203: Based on the data set, obtain the first probability distribution.

[0074] In one implementation manner, when the data set only includes the first battery cell health degree, the data set directly reflects the actual health status of the battery cells, and the first probability distribution is directly constructed based on the first battery cell health degree. This probability distribution can accurately show the distribution of the battery cell health degree in the sample set.

[0075] When the data set includes the health of the first battery cell and other relevant data, in addition to the health of the first battery cell, the data set also includes other variables that may affect the health of the battery cell, such as operating temperature, discharge rate, etc. These additional data enable in-depth analysis of the relationship between the health of the battery cell and these variables through modeling methods. Thus, the health of the battery cells in the entire first battery cell sample set can be calculated. Therefore, the first probability distribution is more comprehensive and universal at this time.

[0076] In this embodiment, for the first probability distribution, this distribution can be corresponding to a single time node or can include multiple different time nodes. For the probability distribution of a single time node, it directly shows the distribution of the health of the battery cell at a specific moment. And the probability distribution including multiple time nodes provides more levels of effective data, which will improve the accuracy of subsequent judgment of the quality of the battery cell health algorithm.

[0077] In this embodiment, if the data set includes not only the health of the first battery cell, then the first probability distribution is obtained through steps S203-1 to S203-4. At this time, the data set includes the values of the health impact variables corresponding to the health of the first battery cell.

[0078] Step S203-1: Based on the data set, obtain the true value model of the battery cell health corresponding to the second battery cell sample set.

[0079] In this embodiment, the true value model of the battery cell health reflects the relationship between the health of the battery cell and the health impact variables.

[0080] In one embodiment, the purpose of the true value model of the battery cell health is to construct a model that can accurately represent the relationship between the health of the battery cell and the key external factors. This model takes into account the influence of external factors on the health of the battery cell to ensure that the battery cell health assessment is closer to the actual usage process.

[0081] In one embodiment, the model implementation is to establish a function model similar to soh = Fx(T, Idch, DOD, Ifch). In this model, the health of the battery cell (SOH) is used as the dependent variable, while the operating temperature (T), discharge rate (Idch), depth of discharge (DOD), and fast charge rate (Ifch) are used as independent variables. This model expresses the relationship between the health of the battery cell and these variables through the mathematical function Fx:

[0082] T (operating temperature): Affects the chemical reaction rate and efficiency of the battery cell, thus affecting its health.

[0083] Idch (discharge rate): Reflects the performance of the battery cell during discharge, and high-rate discharge may accelerate battery aging.

[0084] DOD (Depth of Discharge): It represents the depth of discharge of the battery in a charge-discharge cycle. Deep discharge may reduce the service life of the battery.

[0085] Ifch (Fast Charge Rate): Although fast charging is convenient, it may have a negative impact on battery life.

[0086] In this embodiment, a true value model of battery health is obtained through regression analysis, the least squares method, etc.

[0087] In another embodiment, the true value model of battery health can also be a neural network-based model. The neural network model can handle complex non-linear relationships and is particularly suitable when the influencing factors of cell health are intertwined and complex. By learning a large number of data samples, the neural network can identify deep patterns and trends that affect cell health.

[0088] The neural network model may involve multiple layers of neurons and various connection methods, where each neuron represents a simple function, such as an activation function. The network adjusts the connection weights between neurons to learn and simulate the complex relationship between cell health and health influencing variables.

[0089] In this embodiment, the true value model of battery health can be one model or multiple models, as Figure 4 shown. If the true value model of battery health is multiple models, they are models for different time nodes, that is, a corresponding true value model of battery health is obtained for each time node. If multiple models are obtained, the true value models of battery health are established respectively for the data corresponding to different time nodes in the dataset. It should be noted that if the dataset includes data of multiple different time nodes, only one true value model of battery health can also be generated. Generating one true value model of battery health or multiple true value models of battery health only affects the subsequent accuracy.

[0090] Step S203-2: Obtain the values of health influencing variables corresponding to the cells in the first cell sample set.

[0091] In one embodiment, key parameters that may affect cell health, namely health influencing variables, are collected. These parameters include the operating temperature of the cell, discharge rate, depth of discharge (DOD), fast charge rate, etc.

[0092] Preferably, these values of the health - influencing variables are obtained from samples other than the second battery cell sample set in the first battery cell sample set. In this embodiment, the second battery cell sample set has been tested in detail in a laboratory environment, and the other samples in the first battery cell sample set are put on the market. The data obtained from the market can reflect the performance of the battery cells under actual usage conditions. The purpose of doing this is to ensure that the first probability distribution can more accurately reflect the health status of the battery cells in the real usage scenario.

[0093] In this embodiment, obtaining these values of the health - influencing variables involves monitoring and data recording of the battery cells during actual use. For example, if the battery cells are used in an electric vehicle, then this data may include information such as the battery cell temperature, charge - discharge frequency, and depth under different driving conditions. This data can be collected through the vehicle's battery management system (BMS), which continuously monitors various performance indicators of the battery and records them.

[0094] Step S203 - 3: Based on the true battery health model and the values of the health - influencing variables, obtain the second battery cell health of the battery cells in the first battery cell sample set.

[0095] In this embodiment, the second battery cell health is the health of the first battery cell sample set calculated through the obtained values of the health - influencing variables and the true battery health model, and the second battery cell health is also a true value.

[0096] In one embodiment, the previously established true battery health model and the collected values of the health - influencing variables are used to estimate the second battery cell health of all or part of the battery cells in the first battery cell sample set. This process is based on the characteristic that the battery cells in the first battery cell sample set are similar samples, that is, since the battery cells in the first battery cell sample set are extremely similar in characteristics, the conclusions drawn by the model based on a part of the samples can be effectively extended to the entire sample set.

[0097] In one embodiment, first, the collected values of the health - influencing variables are input into the true battery health model. These variables are considered key factors affecting the battery cell health, and the true battery health model can predict the battery cell health based on the changes of these factors.

[0098] Since the true battery health model has considered these external conditions during construction, the obtained health will be more in line with the actual usage scenarios of the other battery cells in the first battery cell sample set except the second battery cell sample set, making the obtained second health more accurate.

[0099] The advantage of this method is that it allows for generalization from limited experimental data to a broader sample of battery cells, thus providing a comprehensive and accurate overview of the battery cell health. Since the battery cells in the first battery cell sample set are highly similar in production, the second battery cell health obtained by the model can be regarded as an estimate close to the true value. In this embodiment, the second health is considered to be the true value.

[0100] The second battery cell health after the expansion of this step expands the sample, enabling the final first probability distribution to more accurately reflect the distribution of battery cell health in actual use while maintaining its authenticity. This thus lays a solid foundation for subsequent evaluation of different battery cell health algorithms.

[0101] Step S203-4: Obtain a first probability distribution based on the second battery cell health.

[0102] In one embodiment, the first probability distribution provides a comprehensive and accurate view of the health distribution of all or part of the battery cells in the first battery cell sample set.

[0103] In this embodiment, based on the second battery cell health data obtained from the true value model of battery health, a first probability distribution is obtained. The first probability distribution reveals the variations and general characteristics of battery cell health across the entire sample set, such as the central tendency, dispersion, and potential outliers of battery cell health. Preferably, if the samples given in the first probability distribution cover all the battery cells in the first battery cell sample set, then the resulting first probability distribution will more comprehensively reflect the health status of the entire sample set.

[0104] In another embodiment, a first probability distribution is obtained based on the first battery cell health and the second battery cell health.

[0105] In this embodiment, the practice of introducing the second battery cell health to expand the sample greatly enhances the representativeness and accuracy of the first probability distribution. This method allows for the combination of limited experimental data with broader data estimated by the model, thus creating a true battery cell health distribution that reflects both laboratory test results and actual usage.

[0106] In one embodiment, the first probability distribution includes the probability distribution of the battery cell health corresponding to the same time node of the battery cell. Preferably, in another embodiment, the first probability distribution includes the probability distribution at at least two time nodes corresponding to the battery cell.

[0107] The first probability distribution includes the probability distribution of the cell health at the same time node corresponding to the cell, focusing on the cell health at a specific time point. It reflects the health states and distributions of different cells in the cell sample set at that moment. For example, if the health of the cell after one year of use is considered, then this probability distribution will show the health conditions of all cells after one year of use, including those with excellent performance and those with degraded performance. The advantage of this method is that it can eliminate the health differences caused by different usage times, thus providing an accurate snapshot of the cell health at a specific moment.

[0108] The first probability distribution includes the case of the probability distribution of the cell at at least two time nodes corresponding to the cell, which extends the analysis of a single time point and covers the health conditions of the cell at multiple time points. For example, the health conditions of the cell after one year, two years, and three years of use can be analyzed separately. Such a probability distribution reveals the health conditions of the cell at multiple specific time points, strengthens the consideration of time, makes the sample types in subsequent analysis richer, and has higher accuracy.

[0109] In summary, step S203-4 establishes a comprehensive first probability distribution by integrating experimental data and model prediction results, which not only improves the understanding of battery health, but also provides a solid foundation for evaluating and comparing different cell health algorithms.

[0110] Step S30: Based on the first cell sample set and the cell health algorithm, obtain the second probability distribution of the cell health corresponding to the first cell sample set.

[0111] In this embodiment, the second probability distribution is the probability distribution obtained based on the cell health algorithm.

[0112] In one implementation, one or more cell health algorithms are adopted. These algorithms may include algorithms based on empirical models, methods based on statistical analysis, and even more complex machine learning or artificial intelligence techniques. Each algorithm attempts to predict the health condition of the cell based on the historical data and operating parameters of the cell, such as the number of charge and discharge cycles, operating temperature, depth of discharge, and cycle efficiency of the cell.

[0113] After calculating the health of each cell in the first cell sample set using the selected algorithm, the next task is to construct the second probability distribution based on these calculation results. This distribution reflects the distribution of the cell health predicted by the algorithm in the entire sample set, including the central tendency, degree of variation, and potential biases of the predicted values.

[0114] In this implementation, by applying one or more cell health algorithms and constructing the second probability distribution based on the outputs of these algorithms, a basis is provided for subsequent algorithm evaluation and comparison.

[0115] In this embodiment, a specific second probability distribution is obtained through step S301 and step S302, as Figure 3 shown below:

[0116] Step S301: Calculate the cells in the first cell sample set based on the cell health algorithm to obtain the third cell health of the cells in the first cell sample set.

[0117] In one embodiment, the cell health algorithm is applied to calculate the cells in the first cell sample set to estimate their third cell health. This calculation process involves collecting and analyzing the usage data of the cells, and applying a specific health assessment algorithm based on these data.

[0118] In this embodiment, first, the data types and time ranges required for the calculation are determined. For example, if the goal is to calculate the cell health at a specific time node, the usage data of the cell before that time point needs to be collected. These data are determined according to the inputs required by the cell health algorithm.

[0119] Next, the collected data will be used as inputs and fed into the cell health algorithm. These algorithms may be based on different mathematical models, including empirical formulas, statistical models, or complex machine learning algorithms.

[0120] After the algorithm operation is completed, the obtained result is the third cell health of all or part of the cells in the first cell sample set. Preferably, in this embodiment, the third cell health of all the cells in the first cell sample set is obtained. This data reflects the predicted health status of the cells at a specific time node and provides a basis for constructing the second probability distribution.

[0121] Step S302: Obtain the second probability distribution based on the third cell health.

[0122] In one embodiment, the second probability distribution includes the probability distribution of the cell health corresponding to the same time node of the cell. In another embodiment, the second probability distribution includes the probability distribution at at least two time nodes corresponding to the cell.

[0123] In this embodiment, the reason for using the probability distribution is that it can provide more comprehensive information than a single data point or an average value. The probability distribution can show the overall range, central tendency, and degree of variation of the cell health. This method can minimize the influence of individual outliers or cases and provide a more objective and comprehensive assessment of the cell health status.

[0124] In this embodiment, if multiple cell health algorithms are evaluated, each cell health algorithm will generate a second probability distribution.

[0125] Step S40: Obtain the accuracy of the battery cell health algorithm based on the first probability distribution and the second probability distribution

[0126] In one implementation, it is achieved by comparing the first probability distribution and the second probability distribution. These two distributions respectively represent the true value of the battery cell health obtained based on actual measurement and the predicted value of the battery cell health predicted by the battery cell health algorithm

[0127] When evaluating the accuracy, the key lies in comparing the similarity of these two probability distributions. Since these two distributions are generated based on the same battery cell sample set, other influencing factors (such as the differences between battery cells) can be effectively controlled to a minimum. This comparison method ensures the fairness and scientific nature of the evaluation, making the conclusion more persuasive

[0128] During the evaluation process, various statistical methods can be used to compare the similarity of the two probability distributions. For example, metrics such as the Kullback-Leibler Divergence (KL divergence) or the Jensen-Shannon Divergence (JS distance) can be used. These methods evaluate the similarity of the two distributions by calculating the "distance" between them. The smaller the distance, the more similar the two distributions are, and thus the closer the predicted result of the battery cell health algorithm is to the actual measurement (or calculation) result

[0129] For example, if the health distribution predicted by a battery cell health algorithm is highly consistent with the distribution obtained based on experimental data, then it can be considered that this algorithm has a high accuracy. On the contrary, if there are significant differences between the two distributions, it indicates that there may be biases in the battery cell health algorithm when predicting the battery cell health

[0130] In this implementation, the accuracy is obtained through steps S401 - S402, specifically as follows

[0131] Step S401: Determine the distribution type of the first probability distribution, and obtain the accuracy calculation method based on the distribution type

[0132] In this embodiment, the accuracy calculation method is the corresponding calculation method set for the first probability distribution

[0133] In one implementation, statistical hypothesis testing is used to determine whether the first probability distribution conforms to a certain common probability distribution type, and then a suitable statistical method is selected to evaluate the similarity between the probability distributions

[0134] First, determine whether the first probability distribution is a parametric distribution or a non-parametric distribution. This determination is based on statistical analysis of the data, such as testing whether the data conforms to a normal distribution or other known parametric distributions. If the data conforms to a specific parametric distribution model, more accurate statistical methods can be used to evaluate the accuracy of this model.

[0135] If the first probability distribution conforms to a parametric distribution, the Kullback-Leibler Divergence will be selected as the method for evaluating distribution similarity.

[0136] If the first probability distribution does not conform to a parametric distribution, that is, it is a non-parametric distribution, then the Jensen-Shannon Divergence will be used to evaluate the similarity between distributions.

[0137] Step S402: Based on the first probability distribution, the second probability distribution, and the accuracy calculation method, obtain the accuracy.

[0138] In one implementation, according to the accuracy calculation method determined in step S401, such as the Kullback-Leibler Divergence or the Jensen-Shannon Divergence, these two probability distributions are compared. If the distribution of the cell health predicted by the battery health algorithm is very close to the true cell health distribution obtained based on experimental data or model calculations, then this algorithm is considered highly accurate. Conversely, if the two distributions are very different, it means that this cell health algorithm is not applicable to such cells.

[0139] In this implementation, if the obtained first probability distribution and second probability distribution both include the probability distributions of the cell corresponding to at least two time nodes, then steps S403 - S404 are also executed.

[0140] Step S403: For each time node, based on the probability distribution in the first probability distribution corresponding to the same time node and the probability distribution in the second probability distribution corresponding to the same time node, obtain the similarity corresponding to each time node.

[0141] In one implementation, the two probability distributions at each time node are compared to calculate the similarity at each time node.

[0142] In this implementation, the first probability distribution and the second probability distribution at each time node are considered independently. For example, if there are three time nodes (such as one year, two years, and three years after the cell is used), then for each time node, there will be a first probability distribution and a second probability distribution.

[0143] Next, for each time node, compare these two distributions to calculate the similarity. This comparison can be carried out using the methods mentioned in steps S401 - S402, such as KL divergence or JS distance. By calculating these statistical metrics, the similarity degree between the two distributions at each time node can be quantified.

[0144] Step S404: Based on the similarity corresponding to each time node, obtain the accuracy of the battery cell health algorithm.

[0145] In one implementation, integrate the similarities of these time points to evaluate the accuracy of the algorithm during the entire battery life. This comprehensive evaluation can be carried out in various ways. For example, the average value of these similarities can be calculated, or more complex statistical methods can be used to integrate this data to consider the weights or importance of different time nodes.

[0146] Obtain the accuracy of the battery cell health algorithm from the above.

[0147] In this implementation, if the accuracies corresponding to multiple battery cell health algorithms are obtained, then based on the accuracy and preset conditions, obtain the battery cell health algorithm after being screened by the preset conditions.

[0148] Specifically, if the accuracy data corresponding to multiple battery health algorithms are obtained during the evaluation process, the next step is to screen out the optimal battery health algorithm based on these accuracies and preset conditions. The core of this step lies in selecting the most suitable algorithm for a specific application or requirement from multiple candidate algorithms.

[0149] In this implementation, select the battery cell health algorithm with the highest accuracy as the battery cell health algorithm after screening.

[0150] It should be noted that although the above embodiments describe each step in a specific order, those skilled in the art can understand that in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these changes are all within the protection scope of the present invention.

[0151] Furthermore, the present invention also provides a device for evaluating the superiority and inferiority of a battery cell health algorithm.

[0152] Battery cell sample sampling module: Obtain a first battery cell sample set, where the battery cells in the first battery cell sample set are all similar samples to each other;

[0153] True probability distribution module: Based on the first battery cell sample set, obtain the first probability distribution of the battery cell health corresponding to the first battery cell sample set, where the first probability distribution is a probability distribution obtained based on the true value of the battery cell health;

[0154] Algorithm probability distribution module: Based on the first battery cell sample set and the battery cell health algorithm, obtain the second probability distribution of the battery cell health corresponding to the first battery cell sample set, where the second probability distribution is the probability distribution obtained based on the battery cell health algorithm;

[0155] Analysis module: Based on the first probability distribution and the second probability distribution, obtain the accuracy of the battery cell health algorithm.

[0156] In one implementation, the description of the specific implementation functions can refer to steps S10 - 40.

[0157] The above device for evaluating the advantages and disadvantages of the battery cell health algorithm is used to execute Figure 1 the embodiment of the calculation method of the battery cell health shown. The technical principles, the technical problems solved and the technical effects produced by both are similar. Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process and relevant descriptions of the device for evaluating the advantages and disadvantages of the battery cell health algorithm can refer to the content described in the embodiment of the calculation method of the battery cell health, which will not be elaborated here.

[0158] Those skilled in the art can understand that all or part of the processes in the methods of the above embodiments of the present invention can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes a computer program, and the computer program can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable storage medium can include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0159] Furthermore, the present invention also provides a control device. In an embodiment of the control device according to the present invention, the control device includes a processor and a storage device. The storage device can be configured to store a program for executing the method for calculating the cell health degree in the above method embodiment. The processor can be configured to execute the program in the storage device, and the program includes, but is not limited to, the program for executing the method for calculating the cell health degree in the above method embodiment. For the sake of convenience of description, only the parts related to the embodiments of the present invention are shown. For the specific technical details not disclosed, please refer to the method part of the embodiments of the present invention. The control device can be a control device formed by various electronic devices.

[0160] Furthermore, the present invention also provides a computer-readable storage medium. In an embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium can be configured to store a program for executing the method for calculating the cell health degree in the above method embodiment. The program can be loaded and run by a processor to implement the above method for calculating the cell health degree. For the sake of convenience of description, only the parts related to the embodiments of the present invention are shown. For the specific technical details not disclosed, please refer to the method part of the embodiments of the present invention. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiments of the present invention is a non-transitory computer-readable storage medium.

[0161] Furthermore, it should be understood that since the setting of each module is only for illustrating the functional units of the device of the present invention, the physical devices corresponding to these modules can be the processor itself, or a part of the software in the processor, a part of the hardware, or a part of the combination of software and hardware. Therefore, the number of each module in the figure is only illustrative.

[0162] Those skilled in the art can understand that the various modules in the device can be adaptively split or combined. Such splitting or combination of specific modules will not cause the technical solution to deviate from the principle of the present invention. Therefore, the technical solutions after splitting or combination will all fall within the protection scope of the present invention.

[0163] In each embodiment of the present application, the relevant user personal information that may be involved is all processed in strict accordance with the requirements of laws and regulations, following the principles of legality, legitimacy, and necessity, for a reasonable purpose based on the business scenario, and is the personal information actively provided by the user during the use of the product / service or generated due to the use of the product / service, as well as the personal information obtained with the user's authorization.

[0164] The user's personal information processed by the applicant may vary depending on the specific product / service scenario. It shall be subject to the specific scenario of the user's use of the product / service and may involve the user's account information, device information, driving information, vehicle information or other relevant information. The applicant will treat the user's personal information and its processing with a high degree of diligence.

[0165] The applicant attaches great importance to the security of the user's personal information and has taken security protection measures that meet industry standards and are reasonable and feasible to protect the user's information and prevent the personal information from being accessed, publicly disclosed, used, modified, damaged or lost without authorization.

[0166] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for calculating the health degree of an electric core, characterized in that, Comprising: Calculating the cell health degree for the cell using the cell health degree algorithm with an accuracy greater than or equal to the threshold to obtain the cell health degree; Wherein the process of obtaining the accuracy of the cell health degree algorithm includes: Obtaining a first cell sample set, wherein the cells in the first cell sample set are all similar samples to each other; Based on the first cell sample set, obtaining a first probability distribution of the cell health degree corresponding to the first cell sample set, wherein the first probability distribution is a probability distribution based on the true value of the cell health degree; Based on the first cell sample set and the cell health degree algorithm, obtaining a second probability distribution of the cell health degree corresponding to the first cell sample set, wherein the second probability distribution is a probability distribution based on the cell health degree algorithm; Based on the first probability distribution and the second probability distribution, obtaining the accuracy of the cell health degree algorithm.

2. The method for calculating the health degree of an electric core according to claim 1, characterized in that, "Based on the first cell sample set, obtaining a first probability distribution of the cell health degree corresponding to the first cell sample set" includes: Based on the first cell sample set, obtaining a second cell sample set, wherein the second cell sample set is a subset of the first cell sample set; Based on the second cell sample set, obtaining a data set of the cells in the second cell sample set, wherein the data set includes the first cell health degree of the second cell sample set, and the first cell health degree is the true value of the cell health degree; Based on the data set, obtaining the first probability distribution.

3. The method for calculating the health degree of an electric core according to claim 2, characterized in that, "Based on the data set, obtaining the first probability distribution" includes: Based on the data set, obtaining a true value model of the battery health degree corresponding to the second cell sample set, wherein the true value model of the battery health degree reflects the relationship between the cell health degree and the health degree influencing variables, and the data set includes the values of the health degree influencing variables corresponding to the first cell health degree; Obtaining the values of the health degree influencing variables corresponding to the cells in the first cell sample set; Based on the true value model of the battery health degree and the values of the health degree influencing variables, obtaining the second cell health degree of the cells in the first cell sample set; Based on the second cell health degree, obtaining the first probability distribution.

4. The method for calculating the health degree of an electric core according to claim 1, characterized in that, "Based on the first cell sample set and the cell health degree algorithm, obtaining a second probability distribution of the cell health degree corresponding to the first cell sample set" includes: Calculating the cells in the first cell sample set based on the cell health degree algorithm to obtain the third cell health degree of the cells in the first cell sample set; Based on the third cell health degree, obtaining the second probability distribution.

5. The method for calculating the health degree of an electric core according to claim 1, characterized in that, Wherein both the first probability distribution and the second probability distribution include the probability distribution of the cell health degree corresponding to the same time node of the cell.

6. The method for calculating the health degree of an electric core according to claim 1, characterized in that, Wherein both the first probability distribution and the second probability distribution include the probability distributions at at least two time nodes corresponding to the cell.

7. The method for calculating the health degree of an electric core according to any one of claims 1-6, characterized in that, "Based on the first probability distribution and the second probability distribution, obtaining the accuracy of the cell health degree algorithm" includes: Judging the distribution type of the first probability distribution and obtaining an accuracy calculation method based on the distribution type; Based on the first probability distribution, the second probability distribution and the accuracy calculation method, obtaining the accuracy.

8. A device for evaluating the advantages and disadvantages of different algorithms for calculating the health degree of electric cores, characterized in that, Comprising: Cell sample sampling module: Obtain a first cell sample set, where the cells in the first cell sample set are all similar samples to each other; True probability distribution module: Based on the first cell sample set, obtain a first probability distribution of the cell health of the first cell sample set, where the first probability distribution is a probability distribution obtained based on the true value of the cell health; Algorithm probability distribution module: Based on the first cell sample set and the cell health algorithm, obtain a second probability distribution of the cell health of the first cell sample set, where the second probability distribution is a probability distribution obtained based on the cell health algorithm; Analysis module: Based on the first probability distribution and the second probability distribution, obtain the accuracy of the cell health algorithm.

9. A control device, comprising a processor and a storage device, the storage device being adapted to store a plurality of computer programs, characterized in that, The computer program is adapted to be loaded and run by the processor to execute the method for calculating the cell health according to any one of claims 1 to 7.

10. A computer-readable storage medium, in which a plurality of computer programs are stored, characterized in that, The computer program is adapted to be loaded and run by the processor to execute the method for calculating the cell health according to any one of claims 1 to 7.

Images