Energy storage battery life prediction method, apparatus and device, and readable storage medium

By preprocessing the historical data of energy storage batteries and verifying machine learning models, the optimal prediction model is generated, which solves the problem that the energy storage battery life cannot be accurately predicted in the existing technology, and achieves the consideration of complex factors and the accuracy of life. predict.

CN119024208APending Publication Date: 2024-11-26FUJIAN DEKEN ENERGY TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411167276.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The prior art cannot accurately predict the life of energy storage batteries, and it is difficult to fully consider complex factors during battery use, such as ambient temperature, charging and discharging mode and maintenance conditions.

Method used

By obtaining the historical data of the energy storage battery for preprocessing, and dividing it into training sets and test sets of different proportions, multiple machine learning models are called to generate prediction models, and finally obtaining the optimal prediction model for real-time life through test set verification predict.

Benefits of technology

Accurate energy storage battery life prediction based on complex factors is achieved, which improves the accuracy of battery management and maintenance and extends the battery life.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119024208A_ABST
    Figure CN119024208A_ABST
Patent Text Reader

Abstract

The invention provides a life prediction method, device and equipment for an energy storage battery and a readable storage medium, and the method comprises the steps: firstly obtaining historical data of the energy storage battery, and carrying out the preprocessing of the historical data; wherein the historical data comprises charging state data, discharging state data, battery intrinsic data, environment data, special condition data and maintenance record data; then, dividing the preprocessed historical data into a plurality of training sets and test sets with different proportions; the method comprises the steps that firstly, a plurality of training sets are acquired, then, a plurality of machine learning models are called, modeling is carried out based on the plurality of training sets to generate a plurality of prediction models, finally, the corresponding prediction models are verified through a test set to acquire an optimal prediction model, and the optimal prediction model is used for receiving real-time data of an energy storage battery and predicting the service life of the energy storage battery. The problem that the service life of the energy storage battery cannot be accurately predicted based on complex factors in the using process of the energy storage battery in the prior art is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of energy storage battery life prediction, and in particular to a life prediction method, device, equipment and readable storage medium for energy storage batteries. Background Art

[0002] In recent years, energy combustion has been the main source of carbon dioxide emissions, accounting for about 88% of all carbon dioxide emissions. Emissions from the power industry account for about 41% of emissions from the energy industry, and the task of emission reduction is arduous. The key to promoting clean and low-carbon energy transformation is to accelerate the development of new energy. However, new energy has problems such as instability. The large-scale development and utilization of new energy has brought challenges to the safe operation of the power grid. It is urgent to improve the system's flexible adjustment capabilities through measures such as the development of energy storage; As a dynamic and nonlinear electrochemical system, the capacity of energy storage batteries will gradually decay with the increase of charge and discharge times and the extension of continuous use time. Accurate estimation of the state of health (SOH) of batteries is crucial for battery management and maintenance. According to IEEE Standard 1188-1996, when the SOH of energy storage batteries is less than 80%, the batteries need to be replaced.

[0003] At present, accurately predicting the life of energy storage batteries still faces great challenges. Traditional methods mainly rely on simple statistical models or empirical formulas, which are difficult to fully consider the complex factors in the use of batteries, such as ambient temperature, charging and discharging mode, maintenance, etc. These factors will have a significant impact on battery life.

[0004] In view of this, this application is filed. Summary of the invention

[0005] The present invention discloses a life prediction method, device, equipment and readable storage medium for an energy storage battery, aiming to solve the problem that the prior art cannot accurately predict the service life of an energy storage battery based on complex factors during the use of the energy storage battery.

[0006] A first embodiment of the present invention provides a method for predicting the life of an energy storage battery, comprising: Acquire historical data of the energy storage battery and preprocess the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; Dividing the preprocessed historical data into a plurality of training sets and test sets of different proportions; Calling multiple machine learning models and performing modeling based on the multiple training sets to generate multiple prediction models. The corresponding prediction model is verified by the test set to obtain the optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

[0007] Preferably, the preprocessing includes: cleaning abnormal values ​​and missing values ​​in the data set based on the data type, sorting the data in chronological order and converting the time data into a standard time format, and standardizing the historical data, wherein the standardization equation is:

[0008] Z is the standardized data, X is the original data, is the mean of the original data, and std(X) is the standard deviation of the original data.

[0009] Preferably, the preprocessed historical data is divided into a plurality of training sets and test sets of different proportions, specifically: The charging state data and the discharging state data are sorted in chronological order, the battery intrinsic data, the environmental data, and the maintenance record data are combined as feature data, special case data are used as labels, and training and test sets are divided through K-fold cross validation.

[0010] Preferably, the calling of multiple machine learning models and performing modeling based on the multiple training sets to generate multiple prediction models are specifically as follows: S1031, randomly divide the historical data into K parts without repeated sampling, S1032, each time one of the samples is selected as a test set, and the remaining k-1 samples are selected as training sets, and the machine learning model is trained using the training sets; S1033, repeat S1032k times to generate multiple prediction models.

[0011] Preferably, the prediction model corresponding to the test set test is verified by the test set to obtain the optimal prediction model, specifically: Each prediction model calculates the accuracy, recall, and precision of its output through the corresponding test set, and draws a confusion matrix to describe the prediction results of each prediction model, and obtains the optimal prediction model; The expression of the accuracy is: ; The expression of the recall rate is: ; The expression of the accuracy is: ; Among them, TP is the number of positive examples correctly predicted as positive, TN is the number of negative examples correctly predicted as negative, FP is the number of negative examples incorrectly predicted as positive, and FN is the number of positive examples incorrectly predicted as negative.

[0012] Preferably, the maintenance record data is processed using one-hot encoding or N-grams encoding to generate feature data.

[0013] A second embodiment of the present invention provides a life prediction device for an energy storage battery, comprising: A preprocessing unit, used to obtain historical data of the energy storage battery and preprocess the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; A division unit, used for dividing the preprocessed historical data into a plurality of training sets and test sets of different proportions; A prediction model generating unit is used to call multiple machine learning models and perform modeling based on the multiple training sets to generate multiple prediction models. The life prediction unit is used to verify the corresponding prediction model through a test set to obtain an optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

[0014] A third embodiment of the present invention provides a life prediction device for an energy storage battery, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a life prediction method for an energy storage battery as described in any one of the above.

[0015] A fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that a computer program is stored therein, and the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a life prediction method for an energy storage battery as described in any one of the above items.

[0016] Based on the life prediction method, device, equipment and readable storage medium of an energy storage battery provided by the present invention, the historical data of the energy storage battery is first obtained and the historical data is preprocessed; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; then, the preprocessed historical data is divided into a plurality of training sets and test sets of different proportions; then, a plurality of machine learning models are called, and modeling is performed based on the plurality of training sets to generate a plurality of prediction models, and finally, the prediction model corresponding to the test set test is verified by the test set to obtain the optimal prediction model, wherein the optimal prediction model is used to receive the real-time data of the energy storage battery and predict its life. This solves the problem that the prior art cannot accurately predict the service life of the energy storage battery based on the complex factors of the energy storage battery during use. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic flow chart of a method for predicting the life of an energy storage battery provided by the first embodiment of the present invention; Figure 2 It is a schematic diagram of the training process of K-fold cross validation provided by the present invention; Figure 3 It is a schematic diagram of the maintenance position conversion process provided by the present invention; Figure 4 It is a module schematic diagram of a life prediction device for an energy storage battery provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0020] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0022] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0023] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0024] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0025] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] The present invention discloses a life prediction method, device, equipment and readable storage medium for an energy storage battery, aiming to solve the problem that the prior art cannot accurately predict the service life of an energy storage battery based on complex factors during the use of the energy storage battery.

[0027] See also Figure 1 The first embodiment of the present invention provides a life prediction method for an energy storage battery, which can be performed by a life prediction device for an energy storage battery (hereinafter referred to as a prediction device), and in particular, is performed by one or more processors in the prediction device to implement at least the following steps: S101, acquiring historical data of an energy storage battery and preprocessing the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; In this embodiment, the prediction device can be a terminal with data processing capabilities such as a server, workstation, desktop computer, laptop computer, etc., which can communicate with the battery management device in real time. The prediction device can be installed with a corresponding operating system and application software, and the functions required by this embodiment are realized through the combination of the operating system and application software.

[0028] It should be noted that, in this embodiment, the charging status data includes: each charging start time, charging end time, charging efficiency (the ratio of the change value of the battery capacity from the start of charging to the end of charging to the change value of the data recorded by the meter), charging times (the number of times charged per day), charging start temperature (taking the charging instantaneous temperature detection value collected by the temperature sensor device arranged in the battery pack, and taking the average value as the charging start temperature), charging end temperature (taking the end instantaneous temperature detection value collected by the temperature sensor device arranged in the battery pack, and taking the average value as the charging end temperature), charging capacity ( , current integration during charging process, with 10S as sampling interval, to obtain the battery capacity change value during charging); The discharge status data include: the start time of each discharge, the end time of discharge, the number of discharges, the discharge efficiency (the number of discharges per day), the start temperature of discharge (the ratio of the change in battery capacity from the start of discharge to the end of discharge to the data change recorded by the meter), the end temperature of discharge (the temperature detection value at the moment of discharge end collected by the temperature sensor device arranged in the battery pack, and the average value is taken as the end temperature of discharge), and the release capacity (the difference in capacity change before and after discharge is taken as the release capacity).

[0029] The battery intrinsic data include: battery type (mainly cell type: ternary lithium battery, lithium iron phosphate battery, etc.), number of batteries (the number of batteries contained in a battery pack), theoretical capacity (theoretical capacity of the battery when leaving the factory), charge and discharge efficiency (theoretical charge and discharge efficiency of the battery when leaving the factory), charging heat efficiency ℃ / kwh, discharge heat efficiency ℃ / kwh, and self-discharge rate.

[0030] The environmental data include: working temperature (working temperature in the energy storage space, collected every 10 seconds), ambient humidity (ambient humidity collected every ten seconds), remaining area of ​​the space (calculating the remaining area of ​​the space where the energy storage battery pack is located), and ambient wind speed (measuring the wind speed released by the heat dissipation device).

[0031] The special situation data include: battery health SOH.

[0032] Maintenance record data includes: workers' maintenance records for the current year.

[0033] S102, dividing the preprocessed historical data into a plurality of training sets and test sets of different proportions; Specifically, in this embodiment, each charging record is sorted from early to late according to the timestamp to ensure the timing of the data. It can include detailed information such as the charging start time, end time, charging current, voltage, etc. Similarly, each discharge record is sorted according to the timestamp, including information such as the discharge start time, end time, discharge current, voltage, etc. The charging and discharging data are uniformly mapped to the same timeline to ensure the continuity and relevance of the charging and discharging process.

[0034] Furthermore, the battery intrinsic data, environmental data, and maintenance record data are integrated into the feature data table to ensure that each record contains complete feature data. Special case data is added as a label to the feature data table so that each record has corresponding label data. Figure 2 , apply K-fold cross validation, divide the data set into K parts. In each validation, select one of them as the test set and the rest as the training set. In each validation, train the model based on the training set, use the test set to evaluate the model performance, and record the results of each validation. Summarize the results of all K validations, evaluate the average performance of the model, and determine the generalization ability and stability of the model.

[0035] S103, calling multiple machine learning models, and performing modeling based on the multiple training sets to generate multiple prediction models, Specifically, in this embodiment, S1031, first determine the value of K, which can be 5 or 10, depending on the size of the data set and computing resources; use a random number generator to assign a unique index number to each piece of data; evenly distribute the data into K subsets according to the index number, ensuring that the number of samples in each subset is roughly equal; check the data distribution of each subset to ensure that the data distribution between subsets is similar to avoid sampling bias.

[0036] S1032, among the K subsets, select one subset as the test set, and combine the remaining K-1 subsets as the training set; select an appropriate machine learning algorithm, such as decision tree, random forest, support vector machine or neural network; use the training set data to train the selected machine learning model and adjust the model parameters to obtain the best performance; use the test set to evaluate the model performance and calculate evaluation indicators such as accuracy, precision, recall, etc.; record the model performance indicators and model parameters of this round.

[0037] S1033, repeat S1032 K times to generate multiple prediction models: Repeat the cycle K times, each time selecting a different subset as the test set and the rest as the training set; for each cycle, repeat the training and evaluation process in S1032; record the model generated in each cycle and its performance indicators; calculate the average performance indicators of K cross-validations to evaluate the overall performance of the model; analyze the differences among the K models and evaluate the stability and generalization ability of the model; select the model with the best performance, or integrate the K models into an overall model.

[0038] S104, verifying the corresponding prediction model through a test set to obtain an optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

[0039] Each prediction model calculates the accuracy, recall, and precision of its output through the corresponding test set, and draws a confusion matrix to describe the prediction results of each prediction model, and obtains the optimal prediction model; The expression of the accuracy is: ; The expression of the recall rate is: ; The expression of the accuracy is: ; Among them, TP is the number of positive examples correctly predicted as positive, TN is the number of negative examples correctly predicted as negative, FP is the number of negative examples incorrectly predicted as positive, and FN is the number of positive examples incorrectly predicted as negative.

[0040] It should be noted that the accuracy, recall, and precision indicators can be comprehensively considered. According to the specific application scenario, the complexity and computational efficiency of the model can be considered to analyze the performance differences of different models on various indicators. Weighted average or other methods can be used to combine multiple indicators, and the model with the best performance on the target indicator can be selected as the optimal prediction model.

[0041] In a possible embodiment of the present invention, after obtaining real-time data, the system uses a dynamic Bayesian network to adjust the parameters and structure of the prediction model to adapt to the current battery state and environmental changes. This dynamic adjustment mechanism enables the prediction model to respond to fluctuations in external conditions in a timely manner. For example, when the battery load increases or the ambient temperature changes suddenly, the model can automatically adjust its internal parameters to ensure that the deviation between the prediction result and the actual battery life is minimized.

[0042] To further improve the prediction accuracy of the model, the system also implements an online learning mechanism for incremental learning. This mechanism can gradually update the model when new real-time data arrives without retraining the entire model. Incremental learning uses an online stochastic gradient descent algorithm, which enables the model to efficiently absorb the impact of new data while maintaining an understanding of historical data. Through the sliding time window technology, the system dynamically manages the data set for model training, ensuring that the model is mainly trained with the latest and most relevant data, while earlier data is gradually faded or removed to reduce noise interference.

[0043] In a possible embodiment of the present invention, the preprocessing includes: Clean the outliers and missing values ​​in the data set based on the data type. Specifically, outliers can be identified and removed according to the characteristics of each data type. Outliers may be caused by sensor failure, input errors, or emergencies. Statistical methods (such as the 3σ principle, box plot method, etc.) or machine learning algorithms (such as isolation forest, DBSCAN, etc.) can be used to detect outliers. For missing values, different filling methods can be used, such as mean filling, median filling, interpolation, or similarity-based filling methods. The choice of specific methods should be determined according to the characteristics of the data and actual application requirements.

[0044] Sort the data in chronological order and convert the time data into a standard time format. Specifically, sort all data by timestamp to ensure the time continuity and order of the data. Time series sorting can prevent analysis errors caused by data disorder when processing time series data. Convert all time data into a unified standard time format, such as UTC time or local time. Facilitate the compatibility and comparison of data between different systems or different geographical locations.

[0045] And the historical data is standardized, wherein the standardization equation is:

[0046] Z is the standardized data, X is the original data, is the mean of the original data, and std(X) is the standard deviation of the original data.

[0047] It should be noted that the purpose of standardization is to eliminate the dimensional differences between different features and make all feature values ​​on the same scale, thereby improving the training effect of the machine learning model. Through standardization, the influence of each feature on the target variable during model training can be balanced, avoiding some features dominating the model training process due to excessive dimensions.

[0048] In a possible embodiment of the present invention, the maintenance record data is processed using one-hot encoding or N-grams encoding to generate feature data.

[0049] It should be noted that the specific principle is to convert categorical variables (also called discrete features, unordered features) into a format suitable for processing by machine learning algorithms.

[0050] If a set of maintenance records is obtained, it is shown in the following table:

[0051] Computers cannot process text descriptions (categorical variables), so they need to be converted into numerical format before being substituted into computer devices. Taking the maintenance location as an example, the conversion process is as follows: Figure 3 As shown, the numerical matrix is ​​finally substituted into the computer for calculation:

[0052] From the conversion process, we can see that the dimension of the original data is expanded from 1D to 4D. The increase in data volume can also improve the calculation accuracy to a certain extent.

[0053] Please combine Figure 4 The second embodiment of the present invention provides a life prediction device for an energy storage battery, comprising: A preprocessing unit 201 is used to obtain historical data of the energy storage battery and preprocess the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; A division unit 202, used for dividing the pre-processed historical data into a plurality of training sets and test sets of different proportions; The prediction model generating unit 203 is used to call multiple machine learning models and perform modeling based on the multiple training sets to generate multiple prediction models. The life prediction unit 204 is used to verify the corresponding prediction model through a test set to obtain an optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

[0054] A third embodiment of the present invention provides a life prediction device for an energy storage battery, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a life prediction method for an energy storage battery as described in any one of the above.

[0055] A fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that a computer program is stored therein, and the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a life prediction method for an energy storage battery as described in any one of the above items.

[0056] Based on the life prediction method, device, equipment and readable storage medium of an energy storage battery provided by the present invention, the historical data of the energy storage battery is first obtained and the historical data is preprocessed; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; then, the preprocessed historical data is divided into a plurality of training sets and test sets of different proportions; then, a plurality of machine learning models are called, and modeling is performed based on the plurality of training sets to generate a plurality of prediction models, and finally, the prediction model corresponding to the test set test is verified by the test set to obtain the optimal prediction model, wherein the optimal prediction model is used to receive the real-time data of the energy storage battery and predict its life. This solves the problem that the prior art cannot accurately predict the service life of the energy storage battery based on the complex factors of the energy storage battery during use.

[0057] Exemplarily, the computer program described in the third and fourth embodiments of the present invention may be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the device for implementing a life prediction of an energy storage battery. For example, the device described in the second embodiment of the present invention.

[0058] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the life prediction method of the energy storage battery, and uses various interfaces and lines to connect the various parts of the entire method for implementing the life prediction of the energy storage battery.

[0059] The memory can be used to store the computer program and / or module, and the processor realizes various functions of a life prediction method of an energy storage battery by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0060] Wherein, if the implemented module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable 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, computer-readable media do not include electric carrier signals and telecommunication signals.

[0061] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0062] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for predicting the life of an energy storage battery, characterized in that: include: Acquire historical data of the energy storage battery and preprocess the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; Dividing the preprocessed historical data into a plurality of training sets and test sets of different proportions; Calling multiple machine learning models, and performing modeling based on the multiple training sets to generate multiple prediction models; The corresponding prediction model is verified by the test set to obtain the optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

2. A method for predicting the life of an energy storage battery according to claim 1, characterized in that: The preprocessing includes: cleaning the abnormal values ​​and missing values ​​in the data set based on the data type, sorting the data in chronological order and converting the time data into a standard time format, and standardizing the historical data, wherein the standardization equation is: Z is the standardized data, X is the original data, is the mean of the original data, and std(X) is the standard deviation of the original data.

3. The life prediction method of an energy storage battery according to claim 1, characterized in that: The pre-processed historical data is divided into a plurality of training sets and test sets of different proportions, specifically: The charging state data and the discharging state data are sorted in chronological order, the battery intrinsic data, the environmental data, and the maintenance record data are combined as feature data, special case data are used as labels, and training and test sets are divided through K-fold cross validation.

4. The life prediction method of an energy storage battery according to claim 1, characterized in that: The calling of multiple machine learning models and performing modeling based on the multiple training sets to generate multiple prediction models specifically includes: S1031, randomly divide the historical data into K parts without repeated sampling, S1032, each time one of the samples is selected as a test set, and the remaining k-1 samples are selected as training sets, and the machine learning model is trained using the training sets; S1033, repeat S1032k times to generate multiple prediction models.

5. The life prediction method of an energy storage battery according to claim 1, characterized in that: The test set is used to verify the prediction model corresponding to the test set test to obtain the optimal prediction model, specifically: Each prediction model calculates the accuracy, recall, and precision of its output through the corresponding test set, and draws a confusion matrix to describe the prediction results of each prediction model, and obtains the optimal prediction model; The expression of the accuracy is: ; The expression of the recall rate is: ; The expression of the accuracy is: ; Among them, TP is the number of positive examples correctly predicted as positive, TN is the number of negative examples correctly predicted as negative, FP is the number of negative examples incorrectly predicted as positive, and FN is the number of positive examples incorrectly predicted as negative.

6. The life prediction method of an energy storage battery according to claim 1, characterized in that: The maintenance record data is processed using one-hot encoding or N-grams encoding to generate feature data.

7. A life prediction device for an energy storage battery, characterized in that: include: A preprocessing unit, used to obtain historical data of the energy storage battery and preprocess the historical data; wherein the historical data includes charging state data, discharging state data, battery intrinsic data, environmental data, special situation data and maintenance record data; A division unit, used for dividing the preprocessed historical data into a plurality of training sets and test sets of different proportions; A prediction model generating unit, used for calling a plurality of machine learning models, and performing modeling based on the plurality of training sets to generate a plurality of prediction models; The life prediction unit is used to verify the corresponding prediction model through a test set to obtain an optimal prediction model, wherein the optimal prediction model is used to receive real-time data of the energy storage battery and predict its life.

8. A life prediction device for energy storage batteries, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a life prediction method for an energy storage battery as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a life prediction method for an energy storage battery as described in any one of claims 1 to 6.

Citation Information

Cited By

  • Rapid detection system and method for cycle life of lithium ion battery

    CN121049763A

  • Voltage fluctuation suppression method and system for mobile module energy storage

    CN121238571A