Battery capacity prediction method, prediction device and non-volatile readable storage medium

Abstract

The invention discloses a battery capacity prediction method and device and a nonvolatile readable storage medium, the battery capacity prediction method is used for the battery capacity prediction device, and the battery capacity prediction device is configured with a plurality of battery capacity prediction models. The prediction method comprises the following steps: determining a target battery capacity prediction model from a plurality of battery capacity prediction models according to the information of a to-be-detected battery; the predicted capacity of the to-be-detected battery is obtained through the target battery capacity prediction model, and the target battery capacity prediction model takes the real-time process data of the to-be-detected battery as input and takes the predicted capacity of the to-be-detected battery as output. The battery capacity prediction method is short in time consumption, low in energy consumption, high in efficiency and high in prediction precision, so that bad battery cells can be accurately judged, misjudgment is not easy to cause, and the production cost can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of battery testing, and in particular to a battery capacity prediction method, prediction device and non-volatile readable storage medium. Background Art

[0002] Lithium-ion batteries have successfully attracted people's attention due to their advantages of high energy density, long cycle life, environmental friendliness and wide range of applicability. With the significant increase in people's environmental awareness and the pursuit of a green and healthy living environment, new energy vehicles have developed rapidly. Lithium-ion power batteries are widely used and have been widely used in various portable electronic devices and aerospace fields. Therefore, people have also put forward higher requirements on the performance of square lithium-ion power batteries, the preferred environmentally friendly batteries. In order to ensure the battery's endurance, battery capacity detection, i.e. capacity separation, is an important and necessary step. The current common solution is capacity fitting, that is, fully charging the battery and then discharging it to a certain SOC state. After a certain period of time, the OCV is tested. The voltage after the standby period has a certain correlation with the full capacity, and this OCV is used for capacity fitting.

[0003] In the prior art, when the capacity fitting method is used to predict the battery capacity, the entire capacity classification process requires a large amount of equipment and inventory resources, has high production costs, and has disadvantages such as long time consumption, high energy consumption, low efficiency and low prediction accuracy. Therefore, it is impossible to accurately determine defective batteries, which can easily lead to misjudgment. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the first purpose of the present invention is to provide a battery capacity prediction method that is short in time, low in energy consumption, high in efficiency and high in prediction accuracy, so that bad cells can be accurately determined, misjudgment is not likely to occur, and production costs can be reduced.

[0005] The second object of the present invention is to provide a battery capacity prediction device.

[0006] The third object of the present invention is to provide a non-volatile readable storage medium.

[0007] In order to achieve the above-mentioned purpose, an embodiment of the first aspect of the present invention proposes a battery capacity prediction method, which is used for a battery capacity prediction device, wherein the battery capacity prediction device is configured with multiple battery capacity prediction models, and the prediction method comprises: determining a target battery capacity prediction model from the multiple battery capacity prediction models according to the information of the battery to be tested; obtaining the predicted capacity of the battery to be tested through the target battery capacity prediction model, wherein the target battery capacity prediction model takes the real-time process data of the battery to be tested as input and the predicted capacity of the battery to be tested as output.

[0008] According to the battery capacity prediction method proposed in the embodiment of the present invention, the battery capacity prediction models of different types of batteries are pre-configured and saved in the battery capacity prediction device. When predicting the capacity of the battery to be tested, the target battery capacity prediction model that matches the battery to be tested can be directly called, and the real-time process data of the battery to be tested is obtained and input into the prediction model, and the predicted capacity of the battery to be tested can be directly output. The entire prediction process is short in time, low in energy consumption, and high in efficiency. In addition, the real-time process data can be directly obtained from the production line, so that the battery capacity prediction device can realize online prediction of the battery capacity. In addition, by using the trained battery capacity prediction model for prediction, the prediction accuracy is high, so that bad batteries can be accurately judged and it is not easy to cause misjudgment. In addition, in terms of hardware equipment, it can replace the traditional capacity cabinet detection and reduce production costs.

[0009] In some embodiments of the present invention, the information of the battery to be tested includes a battery model.

[0010] In some embodiments of the present invention, the information of the battery to be tested includes at least one of a factory producing the battery to be tested, a production line producing the battery to be tested, and a process name for producing the battery to be tested.

[0011] In some embodiments of the present invention, the multiple battery capacity prediction models correspond to multiple different types of batteries to be tested.

[0012] In some embodiments of the present invention, the real-time process data of the battery to be tested is obtained through a communication network.

[0013] In some embodiments of the present invention, the prediction method further includes: saving real-time process data of the battery to be tested; and optimizing the battery capacity prediction model by machine learning technology based on the saved real-time process data of the battery to be tested.

[0014] In some embodiments of the present invention, the prediction method also includes: completing at least one of the following according to human-computer interaction instructions: startup capacity prediction, process maintenance, data visualization display, system status monitoring, equipment status monitoring, system management and maintenance, parameter setting, and log information query.

[0015] In some embodiments of the present invention, the prediction method further includes: the process maintenance includes updating at least one of the process flow and number of process steps, process duration, capacity standard and change time.

[0016] In some embodiments of the present invention, the prediction method further includes: sending the prediction result of the target battery capacity prediction model to a battery production management system.

[0017] In order to achieve the above-mentioned purpose, the second aspect of the present invention proposes a battery capacity prediction device, comprising: at least one processor; a memory connected to the at least one processor; the memory stores a computer program that can be executed by the at least one processor, and when the at least one processor executes the computer program, it implements the battery capacity prediction method described in any one of the above items.

[0018] According to the battery capacity prediction device proposed in the embodiment of the present invention, when at least one processor executes the computer program stored in the memory to implement the battery capacity prediction method of any of the above embodiments, it can shorten the duration of the entire prediction process, reduce power consumption, improve prediction efficiency, and realize online prediction of battery capacity, and can also improve prediction accuracy, so that bad batteries can be accurately determined and misjudgment is not likely to occur. In addition, in terms of hardware equipment, it can replace traditional capacity cabinet detection and reduce production costs.

[0019] In some embodiments of the present invention, the prediction device also includes: a human-computer interaction device, connected to the at least one processor, for providing a human-computer interaction interface, the human-computer interaction interface is used to receive human-computer interaction instructions, and provide an operation trigger unit according to the human-computer interaction instructions to execute corresponding functions according to user operations.

[0020] In some embodiments of the present invention, the prediction device further includes: a communication device connected to the at least one processor, and configured to receive real-time process data of the battery to be tested.

[0021] In order to achieve the above-mentioned purpose, a third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon, wherein the computer program implements any of the above-mentioned battery capacity prediction methods when executed.

[0022] According to the non-volatile readable storage medium proposed in the embodiment of the present invention, when the stored computer program is executed, by implementing the battery capacity prediction method of any of the above embodiments, the duration of the entire prediction process can be shortened, the power consumption can be reduced, the prediction efficiency can be improved, and the online prediction of the battery capacity can be realized, and the prediction accuracy can be improved, so that the bad battery cells can be accurately determined and misjudgment is not likely to occur. In addition, in terms of hardware equipment, it can replace the traditional capacity cabinet detection and reduce production costs.

[0023] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 A block diagram of a battery capacity prediction device according to an embodiment of the present invention; Figure 2 A flowchart of a method for predicting battery capacity according to an embodiment of the present invention; Figure 3 A block diagram of a device for predicting battery capacity according to another embodiment of the present invention; Figure 4 A flowchart of a method for predicting battery capacity according to another embodiment of the present invention; Figure 5 is a flow chart of a method for predicting battery capacity according to yet another embodiment of the present invention; Figure 6 is a flow chart of a method for predicting battery capacity according to yet another embodiment of the present invention; Figure 7 A block diagram of a device for predicting battery capacity according to another embodiment of the present invention; Figure 8 FIG. 4 is a block diagram of a device for predicting battery capacity according to another embodiment of the present invention.

[0025] Reference numerals: A battery capacity prediction device 100; Processor 10, memory 20, human-computer interaction device 30, communication device 40; Data acquisition unit 1, machine learning modeling unit 2, data prediction processing unit 3, prediction system database 4, human-computer interaction unit 5; Production line unit 11, communication unit 12, data preprocessing module 21, machine learning training module 22, model testing module 23, production line process data processing module 31, prediction system database interface 41, prediction system database server 42, graphical user interface module 51; Device server module 111 , Ethernet module 121 , network security device 122 . DETAILED DESCRIPTION

[0026] Embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. Embodiments of the present invention are described in detail below.

[0027] However, in the current production process, traditional capacity division cabinets are usually used for testing, so the capacity division process usually requires a large amount of equipment and inventory resources, and the production cost is relatively high. In this regard, an embodiment of the present invention proposes a battery capacity prediction method, which can achieve online prediction of battery capacity by acquiring real-time process data of the battery to be tested, and predict according to a pre-established battery capacity prediction model, which can shorten the duration of the entire prediction process, reduce losses, improve prediction efficiency, and improve prediction accuracy, so that bad batteries can be accurately determined without causing misjudgment. In addition, in terms of hardware equipment, it can replace traditional capacity division cabinet detection and reduce production costs.

[0028] In some embodiments of the present invention, a battery capacity prediction method is used for a battery capacity prediction device, and the battery capacity prediction device is configured with multiple battery capacity prediction models. Specifically, the multiple battery capacity prediction models use battery process big data to build a capacity prediction model through an AI algorithm to achieve accurate prediction of the battery cell capacity. The battery capacity prediction model takes the battery process data as input and the predicted capacity of the battery as output. The multiple battery capacity prediction models configured therein can be used to predict the capacity of different batteries on different production lines.

[0029] As for the battery capacity prediction device, from the hardware equipment aspect, the battery capacity prediction device can be a complete set of battery capacity online prediction system. Figure 1 Understanding the battery capacity prediction device of the embodiment of the present invention, Figure 1 The block diagram of a battery capacity prediction device according to an embodiment of the present invention, wherein the battery capacity prediction device 100 may include a data acquisition unit 1 and a main control console, and the main control console is configured with a prediction system server and a computer, etc. From the software aspect, the prediction system server is configured with a software system for predicting battery capacity developed based on machine learning technology. The software system may include a prediction system database 4, a machine learning model unit 2, and a data prediction processing unit 3. The data acquisition unit 1, the prediction system database 4, the machine learning model unit 2, and the data prediction processing unit 3 can exchange data and execute corresponding commands through Ethernet. In addition, the computer can be a human-computer interaction unit 5, which can be used to input human-computer interaction instructions, and can also be used to display the prediction process, production line status information, prediction results, etc., to realize human-computer interaction operations.

[0030] The data acquisition unit 1 is used to obtain the real-time process data of the battery from the production line and store it in the prediction system database 4 of the prediction system server. The machine learning modeling unit 2 is used to train accurate prediction models for various types of batteries. The data prediction processing unit 3 is used to process and predict the real-time process data collected by the data acquisition unit 1, and output the prediction results.

[0031] Reference below Figure 1-Figure 6 A method for predicting battery capacity according to an embodiment of the present invention is described.

[0032] In some embodiments of the present invention, Figure 1-Figure 3 Understand the battery capacity prediction method according to the embodiment of the present invention. Figure 2 A flowchart of a method for predicting battery capacity according to an embodiment of the present invention; Figure 3 FIG. 4 is a block diagram of a device for predicting battery capacity according to another embodiment of the present invention.

[0033] Among them, Figure 2 As shown, the prediction method at least includes step S1 and step S2, which are specifically as follows.

[0034] S1, determining a target battery capacity prediction model from a plurality of battery capacity prediction models according to information of a battery to be tested.

[0035] In some embodiments, the information of the battery to be tested includes the battery model. In other embodiments, multiple battery capacity prediction models correspond to multiple different models of batteries to be tested.

[0036] It is understandable that in actual production and application, different types of batteries may have different production parameters, battery parameters, battery capacity, etc., so it is necessary to train accurate battery capacity prediction models for various types of batteries in advance. Specifically, the battery process big data is used to build capacity prediction models for different types of batteries through machine learning algorithms, and the real-time models are automatically saved to build an online prediction system.

[0037] Specifically, it can be combined Figure 3 Understand the process of establishing and training the battery capacity prediction model according to the embodiment of the present invention.

[0038] The machine learning model unit 2 can retrieve the real-time process data of the historical battery from the prediction system database 4, and train multiple battery capacity prediction models corresponding to different types of batteries, and store the trained prediction models in the prediction system database 4. Figure 3As shown, the machine learning modeling unit 2 includes a data preprocessing module 21, a machine learning training module 22 and a model testing module 23; wherein the data preprocessing module 21 is used to read the collected data such as the battery process from the prediction system database 4, and these data can be historical data. The machine learning training module 22 is used to perform model training and development on the processed data read from the database, and use machine learning technology to deeply explore the characteristic factors related to capacity such as voltage, current, temperature, capacity, etc. of each process, and finally obtain a successfully trained prediction model. The model testing module 23 is used to test and verify the test model obtained by the previous step of machine learning training, determine whether the model is accurate, and finally save the finally successfully trained prediction model in the prediction system database 4.

[0039] In the actual test process, Figure 3 As shown, the data prediction processing unit 3 includes a production line process data processing module 31. The data prediction processing unit 3 retrieves a target battery capacity prediction model corresponding to the model of the battery to be tested from the prediction system database 4 according to the battery signal to be tested.

[0040] S2, obtaining the predicted capacity of the battery to be tested through a target battery capacity prediction model, wherein the target battery capacity prediction model takes the real-time process data of the battery to be tested as input and takes the predicted capacity of the battery to be tested as output.

[0041] Specifically, it can be combined Figure 3 Understand the capacity prediction process of the battery to be tested in the embodiment of the present invention. The prediction system database 4 includes a prediction system database interface 41 and a prediction system database server 42. The data acquisition unit 1 includes a production line unit 11 and a communication unit 12; the production line unit 11 includes a device server module 111; the communication unit 12 includes an Ethernet module 121 and a network security device 122; the data acquisition unit 1 is connected to the prediction system database 4 through the communication unit 12, and the production line process data processing module 31 in the data prediction processing unit 3 extracts the real-time process data corresponding to the model of the battery to be tested from the prediction system database server 42 through the prediction system database interface 41, inputs the target battery capacity prediction model and outputs the prediction result, so as to complete the target battery capacity prediction, and store the prediction result in the prediction system database 4.

[0042] According to the battery capacity prediction method proposed in the embodiment of the present invention, the battery capacity prediction models of different types of batteries are pre-configured and saved in the battery capacity prediction device. When predicting the capacity of the battery to be tested, the target battery capacity prediction model that matches the battery to be tested can be directly called, and the real-time process data of the battery to be tested is obtained and input into the prediction model, and the predicted capacity of the battery to be tested can be directly output. The entire prediction process is short in time, low in energy consumption, and high in efficiency. In addition, the real-time process data can be directly obtained from the production line, so that the battery capacity prediction device can realize online prediction of the battery capacity. In addition, by using the trained battery capacity prediction model for prediction, the prediction accuracy is high, so that bad batteries can be accurately judged and it is not easy to cause misjudgment. In addition, in terms of hardware equipment, it can replace the traditional capacity cabinet detection and reduce production costs.

[0043] Furthermore, the application scenario of the battery capacity prediction method of the embodiment of the present invention can be the initial capacity detection in the capacity division process of the cell detection stage of lithium battery production.

[0044] In some embodiments of the present invention, the real-time process data of the battery to be tested is obtained through a communication network. Figure 3 As shown, the communication unit 12 is used for data interaction between the data acquisition unit 1, the machine learning modeling unit 2, the data prediction processing unit 3 and the human-computer interaction unit 5; the device server module 111, the prediction system server 42, the machine learning modeling unit 2, the data prediction processing unit 3 and the human-computer interaction unit 5 in the data acquisition unit 1 perform data interaction and execute corresponding commands through the Ethernet module 121.

[0045] In some embodiments of the present invention, the information of the battery to be tested includes at least one of a factory producing the battery to be tested, a production line producing the battery to be tested, and a process name for producing the battery to be tested.

[0046] In some embodiments of the present invention, the prediction method also includes: completing at least one of the following according to human-computer interaction instructions: startup capacity prediction, process maintenance, data visualization display, system status monitoring, equipment status monitoring, system management and maintenance, parameter setting, and log information query.

[0047] Among them, it can be combined Figure 3 and Figure 4 Understand step S5 of the embodiment of the present invention. Figure 4 The figure is a flow chart of a method for predicting battery capacity according to another embodiment of the present invention.

[0048] Among them, in some embodiments of the present invention, Figure 3 As shown, the human-computer interaction unit 5 includes a graphical user interface module 51, which is used to interact with the user. The graphical user interface module 51 can be a software interaction interface displayed by a computer of the main control console. Figure 4 As shown, the user can complete at least the operations including startup work, i.e., capacity prediction, process maintenance module, data visualization display, system status monitoring, equipment status monitoring, and system management and maintenance through the graphical user interface module 51. Among them, system status monitoring is used to monitor whether the production line is started and in operation; equipment status monitoring is used to monitor the production equipment corresponding to the collected data, that is, to monitor whether the data acquisition status of the equipment is abnormal. The graphical user interface module includes startup work, system status, data visualization, AI modeling, equipment monitoring, log information, settings, process maintenance and other sections; the startup work section includes parameter setting, start button and startup status monitoring, and the parameter setting section includes selecting factory, production line, process name, whether to start verification, etc.; the system status section can view the system operation status by selecting factory and production line; the data visualization section can view the data chart of the process by selecting factory, production line and process, and the data supports export; the AI ​​modeling section can automatically model by selecting factory, production line and battery; the equipment monitoring section can view whether the data acquisition status is normal by selecting factory, production line and process; the log information section can view the system log information such as startup information, termination information, alarm information, fault information, operation information, etc. by selecting factory and production line; the settings section can configure the relevant personnel permissions and viewing permissions by selecting factory and production line; the process flow and number of steps, process duration, capacity standard and change time can be maintained in the process maintenance section module, and the system automatically saves each change information to adapt to the frequent changes in lithium battery manufacturing process, and avoid inaccurate prediction results due to process changes and changes in feature input.

[0049] Furthermore, the prediction method also includes: process maintenance includes updating at least one of the process flow and the number of process steps, the process duration, the capacity standard and the change time. It is understandable that by setting the process maintenance module 51 in the human-computer interaction unit 5, the process flow and the number of process steps, the process duration, the capacity standard and the change time can be maintained to adapt to the frequent changes in the lithium battery manufacturing process and avoid inaccurate prediction results due to changes in process and changes in feature input.

[0050] In some embodiments of the present invention, Figure 3 and Figure 5 Understand step S5 of the embodiment of the present invention. Figure 5 2 is a flow chart of a method for predicting battery capacity according to yet another embodiment of the present invention. The method for predicting battery capacity further includes step S3 and step S4, which are specifically as follows.

[0051] S3, saving the real-time process data of the battery to be tested.

[0052] like Figure 3As shown, during the actual test process, the data acquisition unit 1 is used to obtain the real-time process data of the battery from the production line, and the real-time process data is stored in the prediction system database server 42 through the Ethernet module 121 and the prediction system database interface 41.

[0053] S4, optimizing the battery capacity prediction model through machine learning technology according to the saved real-time process data of the battery to be tested.

[0054] The stored real-time process data of the battery to be tested can be used as historical process data to continuously optimize the battery capacity prediction model through machine learning technology.

[0055] In some embodiments of the present invention, the prediction method further includes step S5.

[0056] S5, sending the prediction result of the target battery capacity prediction model to the battery production management system.

[0057] Specifically, the prediction results of the target battery capacity prediction model are obtained and stored in the database, which can be used as historical process data, and the battery capacity prediction model is continuously optimized through machine learning technology. In addition, the prediction results are also uploaded to the production management system such as MES through the interface.

[0058] In some embodiments of the present invention, Figure 6 4 is a flowchart of a method for predicting battery capacity according to yet another embodiment of the present invention. The method for predicting battery capacity specifically includes steps S101 - S107 , which are specifically as follows.

[0059] S101, log in to the system.

[0060] S102 , start the work. Open the start work interface, that is, the software interaction interface displayed by the computer, namely the graphical user interface module 51 , and the user can perform the operation by operating the human-computer interaction unit 5 including the graphical user interface module 51 .

[0061] S103, setting parameters, selecting the factory, production line, process name, and whether to start verification. This process can also be used to retrieve the target battery capacity prediction model.

[0062] S104, click Start, and the software system starts running automatically.

[0063] S105, model prediction.

[0064] Specifically, the system collects real-time process data from the device server module 111 and stores it in the prediction system database 4; the data prediction processing unit 3 extracts the real-time process data from the prediction system database 4, aligns and cleans the data, including checking data consistency, processing invalid values ​​and missing values, etc.; and preprocesses the cleaned data to obtain normalized data for input into the prediction model. Finally, the normalized data is input into the target battery capacity prediction model.

[0065] S106, outputting prediction results. The target battery capacity prediction model predicts the capacity of the target battery based on the input real-time process data and outputs the prediction results.

[0066] S107 , the prediction results are stored in the prediction system database and uploaded to a production management system such as MES through the prediction system database interface 41 .

[0067] In some embodiments of the present invention, a battery capacity prediction device is also provided. Figure 7 1 is a block diagram of a battery capacity prediction device according to another embodiment of the present invention, wherein the battery capacity prediction device 100 includes at least one processor 10 and a memory 20 communicatively connected to the at least one processor 10 .

[0068] The memory 20 stores a computer program that can be executed by at least one processor 10 . When the at least one processor 10 executes the computer program, the method for predicting the battery capacity of any of the above embodiments is implemented.

[0069] According to the battery capacity prediction device 100 proposed in the embodiment of the present invention, when at least one processor 10 executes the computer program stored in the memory 20 to implement the battery capacity prediction method of any of the above embodiments, it can shorten the duration of the entire prediction process, reduce power consumption, improve prediction efficiency, and realize online prediction of battery capacity, and can also improve prediction accuracy, so that bad batteries can be accurately determined and misjudgment is not likely to occur. In addition, in terms of hardware equipment, it can replace traditional capacity cabinet detection and reduce production costs.

[0070] In some embodiments of the present invention, Figure 8 This is a block diagram of a battery capacity prediction device according to another embodiment of the present invention. The prediction device 100 also includes a human-computer interaction device 30, which is connected to at least one processor 10 and is used to provide a human-computer interaction interface. The human-computer interaction interface is used to receive human-computer interaction instructions and provide an operation trigger unit according to the human-computer interaction instructions to execute a corresponding function according to a user operation.

[0071] Specifically, the human-machine interaction device 30 can be a computer in the main control console in the above embodiment. The human-machine interaction device 30 can be used to input human-machine interaction instructions, and can also be used to display the prediction process, production line situation information, prediction results, etc., to achieve human-machine interaction operations. Among them, the operation trigger unit can be the startup work, system status, data visualization, AI modeling, equipment monitoring, log information, settings, process maintenance and other sections in the graphical user interface module in the above embodiment.

[0072] In some embodiments of the present invention, Figure 8 As shown, the prediction device 100 also includes a communication device 40, which is connected to at least one processor 10 and is used to receive real-time process data of the battery to be tested. The communication device 40 may include the communication unit 12 in the data acquisition unit 1 in the above embodiment and the prediction system database interface 41 in the prediction system database 4, which is used to realize the transmission of real-time process data of the battery.

[0073] In order to achieve the above-mentioned purpose, a third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon, wherein the computer program implements any of the above-mentioned battery capacity prediction methods when executed.

[0074] According to the non-volatile readable storage medium proposed in the embodiment of the present invention, when the stored computer program is executed, by implementing the battery capacity prediction method of any of the above embodiments, the duration of the entire prediction process can be shortened, the power consumption can be reduced, the prediction efficiency can be improved, and the online prediction of the battery capacity can be realized, and the prediction accuracy can be improved, so that the bad battery cells can be accurately determined and misjudgment is not likely to occur. In addition, in terms of hardware equipment, it can replace the traditional capacity cabinet detection and reduce production costs.

[0075] Other structures and operations of the battery capacity prediction device 100 according to the embodiment of the present invention are known to those skilled in the art and will not be described in detail here.

[0076] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0077] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example.

[0078] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for predicting battery capacity, characterized in that: A device for predicting battery capacity, wherein the device is configured with a plurality of battery capacity prediction models, and the prediction method comprises: Determining a target battery capacity prediction model from the plurality of battery capacity prediction models according to the battery information to be tested; The predicted capacity of the battery to be tested is obtained through the target battery capacity prediction model, wherein the target battery capacity prediction model takes the real-time process data of the battery to be tested as input and takes the predicted capacity of the battery to be tested as output.

2. The prediction method according to claim 1, characterized in that: The information of the battery to be tested includes the battery model.

3. The prediction method according to claim 1, characterized in that: The information of the battery to be tested includes at least one of a factory producing the battery to be tested, a production line producing the battery to be tested, and a process name for producing the battery to be tested.

4. The prediction method according to claim 1 or 2, characterized in that: The multiple battery capacity prediction models correspond to multiple different types of batteries to be tested.

5. The prediction method according to claim 1, characterized in that: The real-time process data of the battery to be tested is obtained through a communication network.

6. The prediction method according to claim 1, characterized in that: The prediction method further comprises: Saving the real-time process data of the battery to be tested; The battery capacity prediction model is optimized using machine learning technology based on the saved real-time process data of the battery to be tested.

7. The prediction method according to claim 1, characterized in that: The prediction method further comprises: Complete at least one of the following according to the human-computer interaction instructions: startup capacity prediction, process maintenance, data visualization, system status monitoring, equipment status monitoring, system management and maintenance, parameter setting, and log information query.

8. The prediction method according to claim 1, characterized in that: The prediction method also includes: the process maintenance includes updating at least one of the process flow and the number of process steps, the process duration, the capacity standard and the change time.

9. The prediction method according to claim 1, characterized in that: The prediction method further comprises: The prediction result of the target battery capacity prediction model is sent to the battery production management system.

10. A battery capacity prediction device, characterized in that: include: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and when the at least one processor executes the computer program, the method for predicting the battery capacity according to any one of claims 1 to 9 is implemented.

11. The prediction device according to claim 10, characterized in that The prediction device also includes: A human-computer interaction device is connected to the at least one processor and is used to provide a human-computer interaction interface, wherein the human-computer interaction interface is used to receive human-computer interaction instructions and provide an operation triggering unit according to the human-computer interaction instructions to execute a corresponding function according to a user operation.

12. The prediction device according to claim 11, characterized in that The prediction device also includes: The communication device is connected to the at least one processor and is used to receive real-time process data of the battery to be tested.

13. A non-volatile readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the method for predicting battery capacity according to any one of claims 1 to 9 is implemented.

Images