Battery analysis method and device based on deep learning
By collecting and preprocessing data in real time during battery charging and using deep learning models to predict battery health status, the problem of insufficient accuracy of battery abnormal warning and health management in the prior art is solved, and more efficient and accurate battery analysis is achieved.
Patent Information
- Application Number
- CN202411880170.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-19
- Publication Date
- 2025-04-15
AI Technical Summary
The existing battery monitoring methods cannot effectively warn of battery abnormalities, and cannot deeply analyze the internal chemical changes of the battery, resulting in insufficient accuracy of battery health management.
The battery analysis method based on deep learning is adopted. By collecting core data in real time during the battery charging process, and after preprocessing, the data is input into the convolutional neural network model and the long-term and short-term memory network model to predict the battery's health status, service life or fault category.
It improves the efficiency and accuracy of battery analysis, and can provide effective early warnings before problems occur, ensuring the accuracy of battery health management.
Smart Images
Figure CN119830097A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of battery analysis technology, and in particular to a battery analysis method and device based on deep learning. Background Art
[0002] With the popularity of electric vehicles and other mobile devices, the continuous increase in battery capacity and power is also accompanied by an increase in safety risks. Traditional battery monitoring methods often lag behind in responding to battery anomalies and cannot provide effective warnings before problems occur. In addition, existing battery health management mostly relies on simple data monitoring, such as voltage, current and other parameters, and cannot conduct in-depth analysis of the complex chemical changes inside the battery, and thus cannot ensure the accuracy of battery health management.
[0003] In summary, how to improve the efficiency and accuracy of battery analysis is an urgent problem that technicians in this field need to solve. Summary of the invention
[0004] In view of this, the present application provides a battery analysis method and device based on deep learning, aiming to improve the efficiency and accuracy of battery analysis.
[0005] In a first aspect, the present application provides a battery analysis method based on deep learning, comprising:
[0006] During the charging process of the battery, core data of the battery is collected in real time;
[0007] Preprocessing the core data to obtain processed core data;
[0008] The processed core data is input into a battery analysis model to obtain a battery analysis result of the battery; the battery analysis model includes a convolutional neural network model and a long short-term memory network model; the convolutional neural network model is used to predict the health status, service life, or fault category of the battery; the long short-term memory network model is used to predict the health status or fault category of the battery.
[0009] Optionally, preprocessing the core data to obtain processed core data includes:
[0010] Using a filtering algorithm to remove noise from the core data to obtain the core data after the noise is removed;
[0011] According to a preset outlier elimination strategy, outliers in the battery temperature data, voltage data, and internal resistance data in the core data after noise removal are eliminated to obtain the core data after outlier elimination;
[0012] The core data after removing the outliers is normalized by using the Min-Max normalization method or the Z-score normalization method to obtain the pre-processed core data.
[0013] Optionally, before inputting the processed core data into a battery analysis model to obtain a battery analysis result of the battery, the method further includes:
[0014] Query charging process data of multiple charging history orders from the distributed database;
[0015] Preprocessing the charging process data to obtain preprocessed charging process data;
[0016] Constructing the convolutional neural network model and the long short-term memory network model;
[0017] The preprocessed charging process data is used to train the convolutional neural network model and the long short-term memory network model.
[0018] Optionally, preprocessing the charging process data to obtain preprocessed charging process data includes:
[0019] Using a filtering algorithm to remove noise from the charging process data to obtain the charging process data after the noise is removed;
[0020] According to a preset outlier elimination strategy, outliers in the battery temperature data, voltage data, and internal resistance data in the charging process data after noise removal are eliminated to obtain the charging process data after outlier elimination;
[0021] The charging process data after the outliers are removed is normalized by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed charging process data.
[0022] Optionally, the using the preprocessed charging process data to train the convolutional neural network model and the long short-term memory network model includes:
[0023] Splitting the preprocessed charging process data into different data sets according to battery type; the data sets include a training set, a validation set, and a test set;
[0024] The convolutional neural network model and the long short-term memory network model are trained using the training set.
[0025] Optionally, after training the convolutional neural network model and the long short-term memory network model using the training set, the method further includes:
[0026] Using the validation set to cross-validate the convolutional neural network model and the long short-term memory network model;
[0027] The accuracy of the convolutional neural network model and the long short-term memory network model is evaluated by the mean absolute error (MAE) indicator and the mean square error (MSE) indicator.
[0028] Optionally, during the charging process of the battery, after collecting the core data of the battery in real time, the method further includes:
[0029] The core data is stored in a distributed database.
[0030] In a second aspect, the present application provides a battery analysis device based on deep learning, comprising:
[0031] A collection module collects the core data of the battery in real time during the charging process of the battery;
[0032] A first preprocessing module, used for preprocessing the core data to obtain processed core data;
[0033] The analysis module is used to input the processed core data into a battery analysis model to obtain a battery analysis result of the battery; the battery analysis model includes a convolutional neural network model and a long short-term memory network model; the convolutional neural network model is used to predict the health status, service life, or fault category of the battery; the long short-term memory network model is used to predict the health status or fault category of the battery.
[0034] Optionally, the first preprocessing module includes:
[0035] A first noise removal unit, configured to remove noise from the core data by using a filtering algorithm to obtain the core data after the noise is removed;
[0036] A first outlier elimination unit is used to eliminate outliers in the battery temperature data, voltage data and internal resistance data in the core data after noise removal according to a preset outlier elimination strategy, so as to obtain the core data after outlier elimination;
[0037] The first normalization unit is used to normalize the core data after the outliers are removed by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed core data.
[0038] Optionally, the device further comprises:
[0039] A query module, used to query the charging process data of multiple charging history orders from a distributed database;
[0040] A second preprocessing module, used for preprocessing the charging process data to obtain preprocessed charging process data;
[0041] A construction module, used to construct the convolutional neural network model and the long short-term memory network model;
[0042] A training module is used to train the convolutional neural network model and the long short-term memory network model using the preprocessed charging process data.
[0043] Optionally, the second preprocessing module includes:
[0044] a second noise removal unit, configured to remove noise from the charging process data by using a filtering algorithm to obtain the charging process data after the noise is removed;
[0045] A second abnormal value elimination unit is used to eliminate abnormal values in the battery temperature data, voltage data and internal resistance data in the charging process data after noise removal according to a preset abnormal value elimination strategy, so as to obtain the charging process data after the abnormal values are eliminated;
[0046] The second normalization unit is used to normalize the charging process data after the outliers are removed by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed charging process data.
[0047] Optionally, the training module includes:
[0048] A splitting unit, used to split the preprocessed charging process data into different data sets according to battery type; the data sets include a training set, a validation set and a test set;
[0049] A training unit is used to train the convolutional neural network model and the long short-term memory network model using the training set.
[0050] Optionally, the device further comprises:
[0051] A cross-validation unit, used to perform cross-validation on the convolutional neural network model and the long short-term memory network model using the validation set;
[0052] An evaluation unit is used to evaluate the accuracy of the convolutional neural network model and the long short-term memory network model through a mean absolute error (MAE) indicator and a mean square error (MSE) indicator.
[0053] Optionally, the device further comprises:
[0054] The storage module is used to store the core data in a distributed database.
[0055] In a third aspect, an embodiment of the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a battery analysis method based on deep learning as described in any one of the implementation methods in the first aspect of the embodiment of the present application is implemented.
[0056] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes a battery analysis method based on deep learning as described in any one of the implementation methods in the first aspect of the embodiment of the present application.
[0057] The present application provides a battery analysis method based on deep learning. When executing the method, the core data of the battery is first collected in real time during the charging process of the battery, and then the core data is preprocessed to obtain the processed core data. Finally, the processed core data is input into the battery analysis model, which includes a convolutional neural network model and a long short-term memory network model. The convolutional neural network model is used to predict the health status, service life, or fault category of the battery, and the long short-term memory network model is used to predict the health status or fault category of the battery. In this way, the battery analysis results can be obtained based on the core data collected in real time. In addition, before the core data is input into the battery analysis model, the core data is preprocessed first, and the data quality of the input model can be improved, thereby improving the accuracy of the battery analysis. In this way, the efficiency and accuracy of battery analysis can be improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0059] Figure 1 A flowchart of a battery analysis method based on deep learning provided in an embodiment of the present application;
[0060] Figure 2 A schematic diagram of the structure of a battery analysis device based on deep learning provided in an embodiment of the present application;
[0061] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0062] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. The present application provides a battery analysis method and device based on deep learning, which is used in the field of battery analysis technology. The above is only an example and does not limit the application field of the method and device name provided in the present application.
[0063] With the popularity of electric vehicles and other mobile devices, the continuous increase in battery capacity and power is also accompanied by an increase in safety risks. Traditional battery monitoring methods often lag behind in responding to battery anomalies and cannot provide effective warnings before problems occur. In addition, existing battery health management mostly relies on simple data monitoring, such as voltage, current and other parameters, and cannot conduct in-depth analysis of the complex chemical changes inside the battery, and thus cannot ensure the accuracy of battery health management.
[0064] After research, the inventor proposed the technical solution of this application. First, during the charging process of the battery, the core data of the battery is collected in real time, and then the core data is preprocessed to obtain the processed core data. Finally, the processed core data is input into the battery analysis model, which includes a convolutional neural network model and a long short-term memory network model. The convolutional neural network model is used to predict the health status, service life, or fault category of the battery, and the long short-term memory network model is used to predict the health status or fault category of the battery. In this way, the battery analysis results can be obtained based on the core data collected in real time. In addition, before the core data is input into the battery analysis model, the core data is preprocessed first, which can improve the data quality of the input model and thus improve the accuracy of the battery analysis. In this way, the efficiency and accuracy of the battery analysis can be improved.
[0065] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application. It should be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.
[0066] See also Figure 1 , Figure 1 A flowchart of a battery analysis method based on deep learning provided in an embodiment of the present application includes:
[0067] S101: During the charging process of the battery, core data of the battery is collected in real time.
[0068] During the charging process of the battery, the core data of the battery is collected in real time, where the core data includes but is not limited to voltage, current, temperature, and state of charge SOC. Specifically, during the charging process, the charging system periodically reads the charging data from the BMS, and the CAN bus protocol or Modbus protocol is used for data collection of the battery monitoring and management system.
[0069] S102: Preprocess the core data to obtain processed core data.
[0070] The collected core data is preprocessed, specifically: using a filtering algorithm to remove noise from the core data to obtain the core data after the noise is removed, wherein the filtering algorithm can be a Kalman filter or a wavelet transform, and then, according to a preset outlier removal strategy, outliers in the battery temperature data, voltage data and internal resistance data in the core data after the noise is removed are removed to obtain the core data after the outliers are removed, and finally, the core data after the outliers are removed is normalized using a Min-Max normalization method or a Z-score normalization method to ensure that the data feature values input into the model are within the same numerical range, which is convenient for model processing and obtains the preprocessed core data.
[0071] S103: Input the processed core data into the battery analysis model to obtain the battery analysis result of the battery.
[0072] The battery analysis model includes a convolutional neural network (CNN) model and a long short-term memory (LSTM) model, wherein the convolutional neural network (CNN) model is used to predict the battery's state of health (SOH), service life (RUL), or fault category; the long short-term memory (LSTM) model is used to predict the battery's state of health (SOH) or fault category. The processed core data is input into the battery analysis model to obtain the final battery analysis results.
[0073] In an embodiment of the present application, the core data of the battery is collected in real time, and then the core data is preprocessed to obtain the processed core data. Finally, the processed core data is input into a battery analysis model, which includes a convolutional neural network model and a long short-term memory network model. The convolutional neural network model is used to predict the health status, service life, or fault category of the battery, and the long short-term memory network model is used to predict the health status or fault category of the battery. In this way, the battery analysis result can be obtained based on the core data collected in real time. In addition, before the core data is input into the battery analysis model, the core data is preprocessed first, which can improve the data quality of the input model and thus improve the accuracy of the battery analysis. In this way, the efficiency and accuracy of the battery analysis can be improved.
[0074] In addition, before executing step S103, it is necessary to first construct and train the convolutional neural network CNN model and the long short-term memory network LSTM model. The specific steps are as follows:
[0075] First, the charging process data of each order is queried from the distributed database, and the queried charging process data is preprocessed. The specific process of the preprocessing is as follows: first, a filtering algorithm is used to remove the noise in the charging process data to obtain the charging process data after the noise is removed, and then according to the preset outlier removal strategy, the outliers in the battery temperature data, voltage data and internal resistance data in the charging process data after the noise is removed are removed to obtain the charging process data after the outliers are removed. Finally, the Min-Max normalization method or the Z-score normalization method is used to normalize the charging process data after the outliers are removed to obtain the preprocessed charging process data.
[0076] Next, the preprocessed charging process data is split into different data sets according to the battery type, where the battery type includes but is not limited to ternary lithium, lithium iron phosphate, etc. Then, the data set is divided into a training set, a validation set, and a test set. In general, the ratio of the number of training sets, validation sets, and test sets is 7:1.5:1.5.
[0077] The convolutional neural network CNN model is constructed as follows:
[0078] Through the convolutional neural network CNN, the convolution kernel can be slid on the input battery data to extract local features. The battery data is a time series of voltage and current. The battery voltage fluctuation, charging current change pattern, temperature fluctuation, etc. may contain key information about the battery health status. The convolution formula is:
[0079] Y(i,j)=(X*W)(i,j)=∑ m ∑ n X(i+m,j+n)·W(m,n);
[0080] Among them, X is the input data, that is, time series data such as voltage and current time series; W is the convolution kernel, which is responsible for extracting feature values from the input data; Y is the output of the convolution operation, usually the feature map after convolution.
[0081] The pooling layer usually follows the convolutional layer to reduce the dimension of the data and reduce the amount of computation while retaining important feature information. The most common pooling method is Max Pooling, which selects the maximum value in the pooling window. The pooling operation formula is:
[0082] Y(i,j)=max m,n∈P(i,j)X(m,n);
[0083] Among them, P(i,j) is the range of the pooling window, X(m,n) is the data in the pooling window, the pooling operation will select the maximum value in the window, and Y(i,j) is the result after pooling. In battery analysis, pooling can help retain key fluctuation features in battery data, such as the peak voltage or rapid changes in current when the battery is charging, thereby reducing the interference of irrelevant information.
[0084] After the convolutional and pooling layers extract features, they are usually passed through one or more fully connected layers for the final classification or regression task. The fully connected layer maps the extracted battery features to a predicted output: battery state of health (SOH), remaining useful life (RUL), or fault category.
[0085] The LSTM model is as follows:
[0086] Since the charging process data is time series data, a long short-term memory network (LSTM) model suitable for processing time series data is selected to capture the long-term dependencies in the time series data for predicting the battery health status and remaining life.
[0087] The core computing unit of LSTM consists of three gates: input gate, forget gate and output gate. The forget gate determines how much information of the memory cell state at the previous moment should be forgotten. Its calculation formula is:
[0088] f t =σ(W f ·[h t-1 ,x t ]+b f );
[0089] where f t is the output of the forget gate, h t-1 is the hidden state at the previous moment, x t is the input at the current moment, that is, the current data of the battery, voltage, current, temperature, etc., W f and b f are the weight and bias of the forget gate, and σ is the sigmoid activation function.
[0090] The input gate determines which information of the current input and the hidden state of the previous moment will be stored in the memory cell. The formula is:
[0091] i t =σ(W i ·[h t-1 ,x t ]+b i);
[0092] Among them, i t is the output of the input gate, W i and b i are the weights and biases of the input gate.
[0093] Next, the memory cell state is updated through the forget gate and the input gate. The formula is:
[0094] C t =f t ·C t-1 +i t tanh(W C ·[h t-1 ,x t ]+b C );
[0095] Among them, C t is the current state of the memory unit, C t-1 is the state of the memory unit at the previous moment, and tanh is the hyperbolic tangent function, which is used to activate the updated memory information.
[0096] The output gate determines the hidden state of the current moment (i.e. the output of the model), which is also passed to the next moment. Its calculation formula is:
[0097] o t =σ(W o ·[h t-1 ,x t ]+b o );
[0098] h t =o t tanh(C t );
[0099] Among them, t is the output of the output gate, h t is the hidden state at the current moment (i.e., the predicted value of battery analysis, which may be a fault prediction or a health state estimation).
[0100] Convolutional neural network (CNN) is used to extract local features of battery timing data (such as rapid changes in voltage, abnormal temperature, etc.) CNN , X CNN As the input passed to the LSTM network, the final LSTM output h t It can be used to predict the battery's state of health (SOH) or failure type.
[0101] Next, the model is trained using the training set. Specifically, supervised learning is used to input multi-dimensional data of battery status, such as voltage, current, temperature, SOC, and the highest single power battery voltage, and output a battery health report. During the training process, the Adam optimizer or RMSprop optimizer can be used to tune the model parameters to reduce the loss function.
[0102] After training the model, you can use the validation set to cross-validate the model to select the best performing model. You can also evaluate the accuracy of the model using the mean absolute error (MAE) and mean square error (MSE) indicators.
[0103] In some embodiments, when the battery analysis model detects battery abnormalities, such as overheating, abnormal increase in internal resistance, etc., an early warning is issued to the user or the management platform.
[0104] In some implementations, the model can be continuously optimized through feedback data to improve the accuracy of battery analysis and make it more generalizable in more scenarios. Specifically, the actual charging effect and health changes of the battery are collected and stored as new training data. Through this feedback mechanism, the training set of the model is continuously expanded and enriched. The model is retrained regularly, and the parameters are tuned in combination with the latest feedback data to improve the model's prediction accuracy of the battery health status and charging strategy.
[0105] The above are some specific implementations of the battery analysis method based on deep learning provided by the embodiments of the present application. Based on this, the present application also provides a corresponding device. The device provided by the embodiments of the present application will be introduced below from the perspective of functional modularization.
[0106] See also Figure 2 , Figure 2 A schematic diagram of a battery analysis device based on deep learning provided in an embodiment of the present application is provided. The battery analysis device based on deep learning 200 includes:
[0107] The acquisition module 210 is used to collect the core data of the battery in real time during the charging process of the battery;
[0108] A first preprocessing module 220, configured to preprocess the core data to obtain processed core data;
[0109] The analysis module 230 is used to input the processed core data into the battery analysis model to obtain the battery analysis results of the battery; the battery analysis model includes a convolutional neural network model and a long short-term memory network model; the convolutional neural network model is used to predict the health status, service life, or fault category of the battery; the long short-term memory network model is used to predict the health status or fault category of the battery.
[0110] Optionally, the first preprocessing module 220 includes:
[0111] A first noise removal unit, configured to remove noise from the core data by using a filtering algorithm to obtain the core data after the noise is removed;
[0112] A first outlier elimination unit is used to eliminate outliers in the battery temperature data, voltage data and internal resistance data in the core data after noise removal according to a preset outlier elimination strategy, so as to obtain the core data after outlier elimination;
[0113] The first normalization unit is used to normalize the core data after the outliers are removed by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed core data.
[0114] Optionally, the device 200 further includes:
[0115] A query module, used to query the charging process data of multiple charging history orders from a distributed database;
[0116] A second preprocessing module, used for preprocessing the charging process data to obtain preprocessed charging process data;
[0117] A construction module, used to construct the convolutional neural network model and the long short-term memory network model;
[0118] A training module is used to train the convolutional neural network model and the long short-term memory network model using the preprocessed charging process data.
[0119] Optionally, the second preprocessing module includes:
[0120] a second noise removal unit, configured to remove noise from the charging process data by using a filtering algorithm to obtain the charging process data after the noise is removed;
[0121] A second abnormal value elimination unit is used to eliminate abnormal values in the battery temperature data, voltage data and internal resistance data in the charging process data after noise removal according to a preset abnormal value elimination strategy, so as to obtain the charging process data after the abnormal values are eliminated;
[0122] The second normalization unit is used to normalize the charging process data after the outliers are removed by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed charging process data.
[0123] Optionally, the training module includes:
[0124] A splitting unit, used to split the preprocessed charging process data into different data sets according to battery type; the data sets include a training set, a validation set and a test set;
[0125] A training unit is used to train the convolutional neural network model and the long short-term memory network model using the training set.
[0126] Optionally, the device 200 further includes:
[0127] A cross-validation unit, used to perform cross-validation on the convolutional neural network model and the long short-term memory network model using the validation set;
[0128] An evaluation unit is used to evaluate the accuracy of the convolutional neural network model and the long short-term memory network model through a mean absolute error (MAE) indicator and a mean square error (MSE) indicator.
[0129] Optionally, the device 200 further includes:
[0130] The storage module is used to store the core data in a distributed database.
[0131] The embodiments of the present application also provide corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.
[0132] like Figure 3 As shown, the computer device 01 is in the form of a general computing device. The components of the computer device 01 may include but are not limited to: one or more processors or processing units 03, a system memory 08, and a bus 04 connecting different system components (including the system memory 08 and the processing unit 03).
[0133] Bus 04 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.
[0134] The computer device 01 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 01, including volatile and non-volatile media, removable and non-removable media.
[0135] The system memory 08 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 09 and / or cache memory 10. The computer device 01 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 11 may be used to read and write non-removable, non-volatile magnetic media ( Figure 3 not shown, usually called a "hard drive"). Although Figure 3 Not shown in the figure, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 04 via one or more data medium interfaces. The memory 08 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.
[0136] A program / utility 12 having a set (at least one) of program modules 13 may be stored, for example, in the memory 08, such program modules 13 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof may include an implementation of a network environment. The program modules 13 generally perform the functions and / or methods of the embodiments described herein.
[0137] The computer device 01 may also communicate with one or more external devices 02 (e.g., keyboard, pointing device, display 07, etc.), one or more devices that enable a user to interact with the computer device 01, and / or any device that enables the computer device 01 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via input / output (I / O) interface 06. Furthermore, the computer device 01 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) via network adapter 05. Figure 3 As shown, the network adapter 05 communicates with other modules of the computer device 01 via the bus 04. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with computer device 01, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0138] The processor unit 03 executes various functional applications and data processing by running programs stored in the system memory 08, such as implementing a battery analysis method based on deep learning provided in an embodiment of the present application.
[0139] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0140] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.
[0141] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.
[0142] The above description is merely an exemplary embodiment of the present application and is not intended to limit the protection scope of the present application.
Claims
1. A battery analysis method based on deep learning, characterized in that: include: During the charging process of the battery, core data of the battery is collected in real time; Preprocessing the core data to obtain processed core data; The processed core data is input into a battery analysis model to obtain a battery analysis result of the battery; the battery analysis model includes a convolutional neural network model and a long short-term memory network model; the convolutional neural network model is used to predict the health status, service life, or fault category of the battery; the long short-term memory network model is used to predict the health status or fault category of the battery.
2. The method according to claim 1, characterized in that The preprocessing of the core data to obtain processed core data includes: Using a filtering algorithm to remove noise from the core data to obtain the core data after the noise is removed; According to a preset outlier elimination strategy, outliers in the battery temperature data, voltage data, and internal resistance data in the core data after noise removal are eliminated to obtain the core data after outlier elimination; The core data after removing the outliers is normalized by using the Min-Max normalization method or the Z-score normalization method to obtain the pre-processed core data.
3. The method according to claim 1, characterized in that Before inputting the processed core data into the battery analysis model to obtain the battery analysis result of the battery, the method further includes: Query charging process data of multiple charging history orders from the distributed database; Preprocessing the charging process data to obtain preprocessed charging process data; Constructing the convolutional neural network model and the long short-term memory network model; The preprocessed charging process data is used to train the convolutional neural network model and the long short-term memory network model.
4. The method according to claim 3, characterized in that The preprocessing of the charging process data to obtain the preprocessed charging process data includes: Using a filtering algorithm to remove noise from the charging process data to obtain the charging process data after the noise is removed; According to a preset outlier elimination strategy, outliers in the battery temperature data, voltage data, and internal resistance data in the charging process data after noise removal are eliminated to obtain the charging process data after outlier elimination; The charging process data after the outliers are removed is normalized by using a Min-Max normalization method or a Z-score normalization method to obtain the pre-processed charging process data.
5. The method according to claim 3, characterized in that: The using the pre-processed charging process data to train the convolutional neural network model and the long short-term memory network model includes: Splitting the preprocessed charging process data into different data sets according to battery type; the data sets include a training set, a validation set, and a test set; The convolutional neural network model and the long short-term memory network model are trained using the training set.
6. The method according to claim 5, characterized in that After the convolutional neural network model and the long short-term memory network model are trained using the training set, the method further includes: Using the validation set to cross-validate the convolutional neural network model and the long short-term memory network model; The accuracy of the convolutional neural network model and the long short-term memory network model is evaluated by the mean absolute error (MAE) indicator and the mean square error (MSE) indicator.
7. The method according to claim 1, characterized in that After collecting the core data of the battery in real time during the charging process of the battery, the method further includes: The core data is stored in a distributed database.
8. A battery analysis device based on deep learning, characterized in that: include: A collection module collects the core data of the battery in real time during the charging process of the battery; A preprocessing module, used for preprocessing the core data to obtain processed core data; The analysis module is used to input the processed core data into a battery analysis model to obtain a battery analysis result of the battery; the battery analysis model includes a convolutional neural network model and a long short-term memory network model; the convolutional neural network model is used to predict the health status, service life, or fault category of the battery; the long short-term memory network model is used to predict the health status or fault category of the battery.
9. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the deep learning-based battery analysis method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the deep learning-based battery analysis method according to any one of claims 1 to 7.