Standardized processing method for multi-type data of lithium battery detection platform

By distinguishing direct acquisition and CAN acquisition methods, the naming and column name of lithium battery detection data tables are standardized, and dirty data is cleaned and data split and merged, the problem of inconsistent data on the detection platform is solved, and data processing efficiency is improved. Accuracy, suitable for performance prediction and simulation model construction of various lithium battery types.

CN119089113APending Publication Date: 2024-12-06安徽国轩新能源汽车科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411112371.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The diversity of different lithium battery detection platforms and internal structures leads to inconsistent naming and listing of detection data, making it difficult to achieve standardized data processing, affecting the efficiency and accuracy of data analysis.

Method used

By distinguishing between direct acquisition and CAN acquisition, the naming and column name of lithium battery detection data tables are standardized, a naming mapping library is established, the naming and column name of data tables is unified, and dirty data is automatically cleaned and split and merged. Suitable for a variety of lithium battery types and testing projects.

Benefits of technology

It improves data processing efficiency and accuracy, facilitates automatic reading of the analysis platform, and is suitable for performance prediction and simulation model construction of different types of batteries, reducing the impact of dirty data on analysis results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119089113A_ABST
    Figure CN119089113A_ABST
Patent Text Reader

Abstract

The invention discloses a lithium battery detection platform multi-type data standardization processing method, which comprises the steps of distinguishing an acquisition mode of a detection platform according to the type of a lithium battery, acquiring lithium battery detection item data by utilizing the detection platform, obtaining a detection data table, standardizing the name of the detection data table, and unifying the column name field format of the detection data table. And obtaining a standardized detection data table with uniform naming specifications and column name formats. According to the lithium battery type, direct collection and CAN collection modes are distinguished, lithium battery detection item data are collected by using the detection platform, a standardized detection data table with a unified naming standard and column name format is obtained, and the standardized detection data table data can be automatically read by an analysis platform; the method greatly improves the processing efficiency and accuracy of the collected data, and lays a foundation for the later prediction of the performances of different types of batteries and the construction of a simulation model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of normalized processing of lithium battery detection data, and in particular to a normalized processing method for multi-type data of a lithium battery detection platform. Background Art

[0002] In recent years, the new energy industry has developed rapidly, and the detection volume of lithium battery modules and battery systems has increased rapidly. However, due to the different internal structures of lithium battery modules and battery system samples, and the diversity of detection platforms, different detection platforms and different internal structures of lithium batteries lead to inconsistent naming and column names of the exported detection data, making it difficult to use these measurement data in a unified and standardized manner.

[0003] Based on the above reasons, a data standardization processing method for multi-type data of a lithium battery testing platform is needed. The different types of detection data of the testing platform can be standardized. The standardized data can be automatically read by the analysis platform, thereby improving the data processing efficiency and accuracy. At the same time, it also lays the foundation for the prediction of different types of battery performance and the construction of simulation models in the later stage.

[0004] Patent document number CN114118961A discloses an automated data analysis system for new energy lithium battery performance testing. By adopting its solution, the analysis efficiency can be improved, human errors and analysis inconsistencies can be avoided, and the quality of battery research and development, trial production and production products can be improved. However, it also has the problem that the data from multiple testing platforms cannot be standardized. Summary of the invention

[0005] The purpose of the present invention is to provide a method for normalizing multi-type data of a lithium battery testing platform, to process multi-type test data tables obtained by the testing platform, to obtain normalized and unified test data, and to facilitate data analysis and processing.

[0006] The purpose of the present invention can be achieved through the following technical solutions: a method for normalizing and processing multi-type data of a lithium battery testing platform, including distinguishing the collection method of the testing platform according to the type of lithium battery, using the testing platform to collect lithium battery testing project data, obtaining a testing data table, standardizing the naming of the testing data table, unifying the column name field format of the testing data table, and obtaining a standardized testing data table with unified naming specifications and column name formats. The standardized testing data table data can be automatically read by the analysis platform, which greatly improves the processing efficiency and accuracy of the collected data.

[0007] Furthermore, the lithium battery types include lithium battery modules and lithium battery systems; wherein the lithium battery system has its own DBC management, achieving full coverage of various types of battery modules and lithium battery systems, and improving the application scope of the detection platform.

[0008] Furthermore, the collection methods include direct collection and CAN collection, wherein: direct collection is to directly collect lithium battery module data using the detection platform; CAN collection is to collect lithium battery system data of the DBC management part using the detection platform, and collect fixed part data not managed by DBC. The detection platform has multiple collection methods, which can be applied to different inspection items, further improving the application scope of the detection platform.

[0009] Furthermore, the test items include: lithium battery capacity energy, cycle life, pulse power, storage class and energy efficiency. The test platform has a variety of test items, covering multiple aspects of lithium battery project testing, and the test data has a wide range of applicability.

[0010] Furthermore, the test data table is named in a standardized manner, and the naming fields include: serial number, test cabinet number, test item name, cooling method, rate, charging method, and current test step name. The standardized naming covers key naming fields. By reading the standardized naming, the key content and application scope of the test data table can be known, and the test data table can be better classified.

[0011] Furthermore, the standardization of the naming of the detection data table includes manual naming, and the manual naming includes:

[0012] According to the test items, manually select the key naming fields, and manually enter the specific content of the key naming fields according to the data in the test data table column name field. The manual naming method has a wide range of applications and is more standardized and accurate. The manual naming method can standardize the naming of lithium battery systems with complex DBC management. After manual naming, it can be saved as a template for direct call when the same type of test is performed again.

[0013] Furthermore, the standardization of the naming of the detection data table also includes automatic naming, and the automatic naming includes:

[0014] When the collection method is direct collection, the column name fields involved in the testing platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When direct collection is performed, the standardized key naming fields are obtained in the standardized naming mapping library based on the testing items, and the specific content of the key naming fields is obtained based on the data in the column name fields of the testing data table.

[0015] When the collection method is CAN collection, the DBC management column name fields of the lithium battery system and the column name fields of the detection platform are pre-summarized and mapped to form a CAN standardized naming mapping library. When CAN collection is performed, the standardized key naming fields are obtained in the CAN standardized naming mapping library based on the detection items, and the specific content of the key naming fields is obtained based on the data in the column name fields of the detection data table.

[0016] Automatic naming can improve naming efficiency and is suitable for rapid and standardized processing of a large number of test data tables. At the same time, automatic naming can more completely obtain the standardized naming of the total lithium battery system including DBC management by distinguishing the collection methods.

[0017] Furthermore, the format of the column name fields in the test data table is unified, including:

[0018] When the collection method is direct collection, the column name fields involved in the testing platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When directly collecting, the column name fields of the test data table obtained by the test are mapped in the standardized naming mapping library according to the test items, and the standardized column name fields of the mapped test data table are obtained;

[0019] When the collection method is CAN collection, the DBC management column name fields of the lithium battery system and the column name fields of the detection platform are pre-summarized and mapped to form a CAN standardized naming mapping library. During CAN collection, based on the test items, the column name fields of the test data table obtained by the test are mapped in the CAN standardized naming mapping library to obtain the standardized column name fields of the mapped test data table.

[0020] After unifying the column name field formats of the test data table, the standardized test data table data can be automatically read by the analysis platform, which greatly improves the efficiency and accuracy of data collection processing.

[0021] Furthermore, it also includes identifying and cleaning dirty data in the detection data table, including the steps of:

[0022] S11, presetting reference values ​​of items in the test data table according to the test items;

[0023] S12, comparing the item value in the acquired test data table with the reference value;

[0024] S13. When the difference between the project value and the reference value exceeds the set threshold, clearing is performed.

[0025] Automatic cleaning of dirty data can standardize the specific content of the detection data table and reduce the impact on the final data analysis.

[0026] Furthermore, it also includes automatically splitting, merging or intercepting the data in each column name field of the acquired detection data table according to the current value and temperature value to obtain the key data required for each detection item. The detection data is automatically split, merged or intercepted through pre-set parameters to obtain the final required key data, which facilitates the joint debugging and testing of different detection platforms and improves the pertinence of data application.

[0027] Beneficial effects of the present invention:

[0028] 1. The present invention distinguishes between direct acquisition and CAN acquisition methods according to the type of lithium battery, uses the detection platform to collect lithium battery detection project data, and obtains a standardized detection data table with unified naming specifications and column name formats. The standardized detection data table data can be automatically read by the analysis platform, which greatly improves the efficiency and accuracy of the collected data processing, and also lays the foundation for the prediction of different types of battery performance and the construction of simulation models in the later stage.

[0029] 2. The present invention pre-plans a naming rule to reflect the key stage information of each detection in the standardized file naming field, which can be compatible with various types of detection items of lithium battery modules and battery systems, and is convenient for the analysis platform to automatically read and obtain key stage information.

[0030] 3. In view of the diversity of internal structures of lithium batteries and the different DBCs of lithium battery systems, which lead to inconsistent column name fields after the test data table is exported, the present invention provides a standardized processing method for column name fields of multiple lithium battery modules and battery systems, which facilitates the acquisition of standardized test data table data and can be automatically read by the analysis platform, thereby greatly improving the processing efficiency and accuracy of collected data.

[0031] 4. In view of the fact that there are many uncontrollable factors in the detection process, such as detection interruption, or pre-processing before and after detection, which may generate some dirty data, the present invention automatically cleans the dirty data, standardizes the specific content of the detection data table, and reduces the impact on the final data analysis.

[0032] 5. Aiming at the problem that different lithium battery type detection items require different key data, the present invention automatically splits, merges or intercepts the detection data through pre-set parameters to obtain the final required key data, facilitates the joint debugging and testing of different detection platforms, and improves the pertinence of data application. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of the structure of a multi-type lithium battery detection platform of the present invention;

[0034] Figure 2 A schematic diagram of a process for normalizing multi-type data of a lithium battery testing platform according to the present invention;

[0035] Figure 3 This is a schematic diagram of the naming format of the detection data table of the present invention;

[0036] Figure 4 This is a schematic diagram of the format of the column name fields in the detection data table of the present invention;

[0037] Figure 5 It is a schematic diagram of the detection data in the detection data table of the present invention. DETAILED DESCRIPTION

[0038] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0039] The present invention discloses a method for normalizing multi-type data of a lithium battery detection platform. The detection platform structure used by the method of the present invention is as follows: Figure 1 As shown, the detection platform includes components such as a control computer, a power cabinet and an environmental chamber.

[0040] Among them, the environmental chamber is connected to the power cabinet with power lines and communication lines, the control computer and the power cabinet are connected with communication lines, the environmental chamber is connected to the lithium battery (sample) for data collection of lithium battery detection projects, and the collected data is transmitted to the control computer through the communication line for processing. The power line and communication line connect the control computer, power cabinet and environmental chamber together to ensure that the power and communication for data collection of lithium battery (sample) detection projects remain normal.

[0041] Figure 1 The connection method of the detection platform includes but is not limited to the connection method shown in the figure. Other connection methods can meet the data collection of lithium battery detection projects.

[0042] The types of lithium batteries to which this method is applied include: lithium battery modules and lithium battery systems, wherein the lithium battery system has its own DBC management, while the lithium battery module does not have separate DBC management.

[0043] Depending on the type of lithium battery, the collection method is divided into direct collection and CAN collection. Direct collection is a way to automatically collect data through the management system that comes with the testing platform. Direct collection uses the testing platform to directly collect lithium battery module data. After collection, the fixed naming fields that come with the management system that comes with the testing platform are used. The naming can be preset through the testing platform system and is relatively standardized.

[0044] CAN acquisition method: CAN acquisition uses the detection platform to collect the lithium battery system data of the DBC management part, and collects the fixed part data that is not managed by DBC, including the DBC management part and the fixed part. Due to the various forms of DBC in the lithium battery system, different types of lithium battery systems have different DBCs, and different internal structures of lithium battery systems, the format and naming of the exported detection data table are not unified, making it difficult to use these data in a unified and standardized manner.

[0045] In view of the above problems, the present invention provides a lithium battery module and a lithium battery system, such as Figure 2As shown, the two methods of direct acquisition and CAN acquisition of the detection platform are distinguished. The detection platform is used to collect data of lithium battery modules and lithium battery system detection items, obtain detection data tables, unify the field formats of the detection data tables, and standardize the naming of the detection data tables, so that a standardized detection data table with unified naming specifications and column name formats can be obtained, and the detection data of different lithium battery types can be standardized. The standardized detection data can be automatically read by the analysis platform, which greatly improves the data processing efficiency and accuracy, and also lays the foundation for the prediction of different types of battery performance and the construction of simulation models in the later stage.

[0046] The testing items for lithium batteries may include lithium battery capacity energy, cycle life, pulse power, storage type and energy efficiency, etc.

[0047] Among them, the naming of the test data table is standardized. According to different test items, the type of lithium battery to be collected can be obtained, the corresponding collection method can be determined, the test items can be selected, the naming rules can be prepared in advance, and the test data table can be named in a standardized manner. The naming fields can be determined by summarizing the test data table and selecting multiple key naming fields as the naming.

[0048] For example Figure 3 As shown, the naming fields may include key information such as serial number, test cabinet number, test item name, cooling method, rate, charging method, and current test step name.

[0049] There are two ways to standardize the naming of the test data table: manual naming and automatic naming:

[0050] Manual naming is mainly used for complex lithium battery systems, whose built-in DBC management cannot be automatically identified, so manual naming standards are used.

[0051] When using the manual method, you can manually select the key named fields based on the test items, and then manually enter the specific content of the key named fields based on the data in the column name field of the test data table.

[0052] When using automatic naming, you can do it in the following ways:

[0053] When the collection method is direct collection, the column name fields involved in the detection platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When directly collecting, the standardized key naming fields are obtained in the standardized naming mapping library based on the type of detection project, and the specific content of the key naming fields is obtained based on the data in the column name fields of the detection data table.

[0054] When the collection method is CAN collection, CAN collection uses the detection platform to collect the lithium battery system data of the DBC management part, and collects the fixed part data that is not managed by DBC; the fixed part data can be directly mapped and obtained through the standardized naming mapping library; the DBC management part can collect and summarize the DBC management column name fields of the lithium battery system in advance, and then map them with the detection platform column name fields, so that the DBC management part forms a CAN standardized naming mapping library. During CAN collection, the standardized key naming fields are obtained in the CAN standardized naming mapping library according to the detection items, and the specific content of the key naming fields is obtained according to the data in the column name fields of the detection data table.

[0055] When using CAN collection, DBC manages the collection of some column name fields, which can be manually matched with the pre-collected standard column name fields. After matching, they are changed to standardized column name fields. After matching, the template can be saved, and the template can be directly applied to the same project in the future to standardize the column name fields.

[0056] The standardized naming format is as follows Figure 3 As shown in the figure, when the test item is capacity energy, the serial number, test cabinet number, test item name, cooling method, rate, charging method and current test step name can be selected as the key naming fields, and these naming fields come from the mapping library collected in advance. The specific content in the rate naming field comes from the data in the column name field of the test data table, including (0.33C, 1C, 0.33C, 1C) and other data.

[0057] like Figure 4 As shown, the format of the column name fields of the test data table is unified. According to the lithium battery test items, the standardized column name fields of the test data table corresponding to the test items can be determined, and the column name fields of the test data table can be unified based on the standardized column name fields, so that different test data tables are standardized using the same column name fields.

[0058] When the collection method is direct collection, the column name fields involved in the detection platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When directly collecting, the column name fields of the detection data table obtained by the detection are mapped in the standardized naming mapping library according to the detection items to obtain the standardized column name fields of the mapped detection data table.

[0059] When the collection method is CAN collection, the DBC management column name fields of the lithium battery system and the column name fields of the detection platform are pre-summarized and mapped to form a CAN standardized naming mapping library. During CAN collection, based on the test items, the column name fields of the test data table obtained by the test are mapped in the CAN standardized naming mapping library to obtain the standardized column name fields of the mapped test data table.

[0060] For example: Figure 4 The first column on the left is the columns column, and the other columns are the column name fields of the test data table exported for different lithium battery types. When the test table is imported, the column name field information in the table is automatically captured to achieve automatic adjustment. The columns column summarizes the column name fields of each test data table and standardizes them. For example, when performing the Nebula-PACK standardized XZ project test, the test data table "Cycle" maps the "TotCycle" column name field item.

[0061] Due to the different internal structures of lithium batteries, the lithium battery system collects voltage and temperature information through its internal DBC system. The column names of the cell voltage and temperature parts of the exported data are not fixed, and the detection platform system cannot recognize them. You can adjust the columns and manually match the temperature and voltage information with the standard column name fields on the left. After saving, the header will be automatically replaced with the standard column name on the left.

[0062] Furthermore, during the lithium battery testing process, some dirty data may be generated due to many uncontrollable factors or pre- and post-test processing of the lithium battery. In order to prevent the dirty data from affecting the test results statistically analyzed by the analysis platform, the present invention has an automatic dirty data cleaning function for different lithium battery type test data.

[0063] Identify and clean dirty data in the test data table, including the following steps:

[0064] S11, presetting reference values ​​of items in the test data table according to the test items;

[0065] S12, comparing the item value in the acquired test data table with the reference value;

[0066] S13. When the difference between the project value and the reference value exceeds the set threshold, clearing is performed.

[0067] For example: judge the capacity value data according to the cycle life test item, compare the capacity item value in the test data table with the reference rated capacity value, and pre-set the judgment threshold. For example, in the cycle life test item, if the difference between the adjacent weeks of the capacity item value and the reference rated capacity value is less than 2, which is lower than the given judgment threshold range, it can be judged as dirty data and cleaned. By cleaning the dirty data, the influence of dirty data on the test results during the test process can be reduced.

[0068] Furthermore, during the lithium battery testing process, the test data in the test data table can be automatically split, merged or intercepted according to pre-set parameters to obtain the required key data.

[0069] There are many types of testing items for lithium battery modules and lithium battery systems. Different testing items require different key data. In recent years, different testing platforms have begun to realize the joint debugging function. Therefore, a test data table may include charging and discharging data at different rate points and different temperature points. The data are mixed together, resulting in the testing platform being unable to distinguish the data and unable to read the required complete data.

[0070] In view of the above problem, according to different detection items, the judgment condition is preset, for example, the judgment condition adopts the current value or the temperature value.

[0071] The charge and discharge process and the charge and discharge rate can be determined based on the current value. When it is detected that there are multiple complete charge and discharge processes in a test data table and the charge or discharge rate points are different, it is determined to be charging and discharging at different rate points, and the test data table will be split into multiple files according to the different rate points.

[0072] The detection temperature point can also be determined according to the charge and discharge starting temperature value. The charge and discharge starting temperature value range can be pre-set, generally ±2°C of the target value. When multiple charge and discharge starting temperatures are detected in a file, the file is automatically split into data of multiple temperature points according to the charge or discharge starting temperature. Each file includes charge and discharge data of one temperature point.

[0073] For example: Figure 5 As shown, a detection data table contains the above three sets of charge and discharge data at the same time. The discharge current value is set to 1C (100±2A), and the discharge starting temperature value is 25±2℃. The platform will first search according to the current value and split the file into two files, namely file number 1 and number 2. Then it will search according to the temperature value and split the file number 1 into two files, namely number 1 and number 3, and finally form three files.

[0074] Furthermore, the present invention can also judge the charging and discharging stage according to the current value, automatically intercept and analyze the required data according to different detection items, and finally output a standardized file.

[0075] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for normalizing multi-type data of a lithium battery testing platform, characterized in that: include: According to the type of lithium battery, distinguish the collection method of the testing platform, use the testing platform to collect lithium battery testing project data, obtain the testing data table, standardize the naming of the testing data table, unify the column name field format of the testing data table, and obtain a standardized testing data table with unified naming specifications and column name formats.

2. A method for normalizing multi-type data of a lithium battery testing platform according to claim 1, characterized in that: The lithium battery types include lithium battery modules and lithium battery systems; Among them, the lithium battery system comes with DBC management.

3. A method for normalizing multi-type data of a lithium battery testing platform according to claim 2, characterized in that: The collection methods include direct collection and CAN collection, where: Direct acquisition is to use the detection platform to directly collect lithium battery module data; CAN acquisition uses the detection platform to collect the lithium battery system data of the DBC management part, and collects the fixed part data that is not managed by DBC.

4. A method for normalizing multi-type data of a lithium battery testing platform according to claim 3, characterized in that: The test items include: lithium battery capacity energy, cycle life, pulse power, storage class and energy efficiency.

5. A method for normalizing multi-type data of a lithium battery testing platform according to claim 4, characterized in that: The test data table is named in a standardized manner, and the naming fields include: Serial number, test cabinet number, test item name, cooling method, rate, charging method and current test step name.

6. A method for normalizing multi-type data of a lithium battery testing platform according to claim 5, characterized in that: Standardizing the naming of the test data table includes manual naming, which includes: Manually select key named fields based on the test items, and manually enter the specific content of the key named fields based on the data in the column name fields of the test data table.

7. A method for normalizing multi-type data of a lithium battery testing platform according to claim 5, characterized in that: The standardization of the naming of the test data table also includes automatic naming, which includes: When the collection method is direct collection, the column name fields involved in the testing platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When directly collecting, the standardized key naming fields are obtained in the standardized naming mapping library based on the test items, and the specific content of the key naming fields is obtained based on the data in the column name fields of the test data table; When the collection method is CAN collection, the DBC management column name fields of the lithium battery system and the column name fields of the detection platform are pre-summarized and mapped to form a CAN standardized naming mapping library. When CAN collection is performed, the standardized key naming fields are obtained in the CAN standardized naming mapping library based on the detection items, and the specific content of the key naming fields is obtained based on the data in the column name fields of the detection data table.

8. The method for normalizing multi-type data of a lithium battery testing platform according to claim 5, characterized in that: Unify the format of the column name fields in the test data table, including: When the collection method is direct collection, the column name fields involved in the testing platform are summarized in advance, and a standardized naming mapping library is established based on the summary. When directly collecting, the column name fields of the test data table obtained by the test are mapped in the standardized naming mapping library according to the test items, and the standardized column name fields of the mapped test data table are obtained; When the collection method is CAN collection, the DBC management column name fields of the lithium battery system and the column name fields of the detection platform are pre-summarized and mapped to form a CAN standardized naming mapping library. During CAN collection, based on the test items, the column name fields of the test data table obtained by the test are mapped in the CAN standardized naming mapping library to obtain the standardized column name fields of the mapped test data table.

9. A method for normalizing multi-type data of a lithium battery testing platform according to any one of claims 1 to 8, characterized in that: It also includes the identification and cleaning of dirty data in the detection data table, including the following steps: S11, presetting reference values ​​of items in the test data table according to the test items; S12, comparing the item value in the acquired test data table with the reference value; S13. When the difference between the project value and the reference value exceeds the set threshold, clearing is performed.

10. A method for normalizing multi-type data of a lithium battery testing platform according to any one of claims 1 to 8, characterized in that: Also includes, According to the current value and temperature value, the data in each column name field of the acquired test data table is automatically split, merged or intercepted to obtain the key data required for each test item.