Battery formation process anomaly detection method, device, equipment and medium

Abstract

The invention discloses a battery formation process anomaly detection method, device and equipment and a medium in the technical field of battery production, and the detection method comprises the steps: obtaining the formation parameters of each battery in the battery formation process, and carrying out the preprocessing of the formation parameters; calculating a formation parameter mean value of each working section in the battery formation process according to the preprocessed formation parameters; and performing anomaly detection by adopting an isolated forest algorithm according to the formation parameter mean value of each workshop section. According to the method, the formation parameters are collected, preprocessed and averaged, so that the reliability of the formation parameters is improved, and the accuracy of subsequent anomaly detection processing is ensured. The processed formation parameters are subjected to anomaly detection through the isolated forest algorithm, compared with a traditional scheme, the method is not influenced by subjective experience of workers, meanwhile, shutdown inspection is not needed, and the production efficiency is considered while the accuracy is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of battery production, and in particular to a method, device, equipment and medium for detecting abnormality in a battery formation process. Background Art

[0002] In the battery formation process, the battery after injection is charged for the first time to activate the active materials in the battery to form SEI film to ensure battery performance and safety. During the formation process, poor formation may occur due to problems with the battery itself, or due to blockage of the nozzle of the formation cabinet, or due to problems with the formation voltage and capacity detection device, resulting in misjudgment of poor formation. These problems greatly affect the qualified rate and defective rate of battery production. At present, the main methods for abnormal detection of the formation process in the industry are: 1. Poor formation of a single battery: Calculate the average voltage of the voltage curve of all cells in the formation cabinet at a specific time point in a specific step, and determine the battery corresponding to the voltage value far from the voltage average as abnormal.

[0003] 2. Batch battery formation is poor: stop the machine and manually check whether there is a nozzle blockage, clean and replace the nozzle.

[0004] In the above abnormality detection method, the specific time point and the voltage mean distance are set by the staff based on experience. Different batteries have different formation states, and it is difficult to accurately determine the specific time point and the voltage mean distance. Manual inspection during shutdown seriously affects production efficiency. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method, device, equipment and medium for detecting abnormalities in a battery formation process, so as to solve the technical problems that the existing abnormality detection methods are affected by human subjective experience on the one hand and the accuracy is difficult to guarantee, and on the other hand the production efficiency is seriously affected by the shutdown inspection.

[0006] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for detecting abnormality in a battery formation process, comprising: Acquiring formation parameters of each battery during the battery formation process, and preprocessing the formation parameters; Calculate the mean value of the formation parameters of each section in the battery formation process according to the pre-treated formation parameters; The isolation forest algorithm is used to detect anomalies based on the mean values ​​of the formation parameters of each process section.

[0007] In the above scheme, the reliability of the formation parameters is improved by collecting, preprocessing and averaging the formation parameters, ensuring the accuracy of subsequent abnormality detection. The isolation forest algorithm is used to detect abnormalities in the processed formation parameters. Compared with traditional schemes, it is not affected by human subjective experience and does not require shutdown inspection. It ensures accuracy while taking into account production efficiency.

[0008] Optionally, the formation parameters include formation voltage and formation capacity.

[0009] The formation parameters include a lot of specific data. In the present invention, the formation voltage and formation capacity are comprehensively selected according to factors such as abnormalities in the battery itself, blockage of the nozzle of the formation cabinet, or abnormalities in the detection equipment, so as to obtain the required abnormal detection results, thereby avoiding the selection of too much data, resulting in large calculation amount and slow detection efficiency.

[0010] Optionally, the preprocessing of the formation parameters includes deduplication, removal of gaps and standardization of the formation parameters.

[0011] The above preprocessing process helps to improve data quality and provide a reliable foundation for subsequent data analysis and mining. Data deduplication refers to identifying and deleting duplicate data records from a data set to ensure the uniqueness of the data. Data denulling refers to deleting null values ​​or missing values ​​in a data set. Null values ​​may be caused by incomplete data entry, data loss, etc. The existence of null values ​​will affect the accuracy of data analysis and mining, so they need to be processed. Data standardization refers to the process of converting data of different formats, structures or meanings into a unified standard format. Data standardization helps to improve the comparability and analyzability of data.

[0012] Optionally, the performing anomaly detection using an isolation forest algorithm according to the mean value of the formation parameter of each process section includes: The isolated Senli algorithm is used to divide the mean value of the formation parameter of each section corresponding to each battery, and the number of divisions corresponding to each battery is counted; A battery whose number of cuts is greater than a number threshold is determined to be abnormal, and a battery whose number of cuts is less than or equal to the number threshold is determined to be not abnormal.

[0013] Isolation Forest (iForest) is a machine learning algorithm for anomaly detection. It is based on a tree structure and can quickly segment data. It does not require labeled training data and is suitable for unsupervised learning scenarios. It has a linear time complexity, which makes it very efficient when processing large-scale data. It has high accuracy in anomaly detection and can accurately identify abnormal data points.

[0014] Optionally, the adopting of the isolated Senli algorithm to divide the mean value of the formation parameter of each section corresponding to each battery includes: Use 0 as the base point and repeat the following steps until no more split points can be selected: Calculating the formation parameter metric of each battery according to the formation parameter mean of each section, and determining the metric range according to the maximum and minimum values ​​of the formation parameter metric; A formation parameter of a battery is randomly selected as a split point, and the formation parameter measurement of the selected battery is within the measurement range, and the split point is used as a base point.

[0015] The Isolation Forest algorithm is based on the assumption that outliers are a minority in the data. Their distribution in the feature space is different from that of normal data points, and they are usually far away from most data points. The algorithm detects outliers by constructing multiple isolation trees. Each isolation tree is a binary tree structure that recursively generates partitions of the data set by randomly selecting features and feature split values ​​until the stopping condition is met (such as the number of data points in the subset is less than a certain threshold).

[0016] Optionally, the transformation parameter metric is represented by Euclidean distance; When 0 is taken as the base point, the expression is: ; In the formula, When 0 is the base point The formation parameter measurement of a battery, For the The battery corresponding to The mean value of the formation parameters of each process section, , is the number of work sections; When the split point is used as the base point, the expression is: ; In the formula, When the split point is used as the base point The formation parameter measurement of a battery, The first The mean value of the formation parameters of each process section.

[0017] Optionally, for a single battery, if both the formation voltage and the formation capacity are judged to be abnormal, then the battery is determined to have a formation abnormality; If the number of batteries with abnormal formation on the same tray exceeds the quantity threshold, it is determined that the nozzle corresponding to the tray is abnormal; For a single battery, if it is determined that there is an abnormality based on the formation voltage, and it is determined that there is no abnormality based on the formation capacity, then it is determined that the battery has a formation voltage abnormality; If the number of batteries with abnormal formation voltage on the same tray exceeds the quantity threshold, it is determined that the voltage detection device corresponding to the tray is abnormal; For a single battery, if it is determined that there is no abnormality based on the formation voltage, and that there is an abnormality based on the formation capacity, then it is determined that the battery has a formation capacity abnormality; If the number of batteries with abnormal formation capacity on the same tray exceeds the quantity threshold, it is determined that the capacity detection device corresponding to the tray is abnormal; For a single battery, if it is determined that there is no abnormality based on both the formation voltage and the formation capacity, it is determined that there is no formation abnormality in the battery.

[0018] In the above scheme, the single battery and the batteries on the entire tray are analyzed separately according to the formation voltage and formation capacity, and the abnormality judgment is made on the battery itself, the voltage detection equipment, the capacity detection equipment, and the nozzle of the formation cabinet respectively, which reduces the production waste of manual inspection due to shutdown and has excellent detection effect.

[0019] In a second aspect, the present invention provides a battery formation process abnormality detection device, comprising: A parameter acquisition processing module is configured to acquire the formation parameters of each battery in the battery formation process and pre-process the formation parameters; A formation parameter calculation module is configured to calculate the mean value of the formation parameters of each section in the battery formation process according to the pre-processed formation parameters; The abnormality detection and judgment module is configured to perform abnormality detection using an isolation forest algorithm based on the mean value of the formation parameters of each process section.

[0020] In a third aspect, the present invention provides an electronic device, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0021] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0022] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a battery formation process abnormality detection method, device, equipment and medium. The detection method improves the reliability of the formation parameters by collecting, preprocessing and averaging the formation parameters, and ensures the accuracy of subsequent abnormality detection processing. The processed formation parameters are detected by an isolation forest algorithm. Compared with the traditional scheme, it is not affected by human subjective experience, and there is no need to stop the machine for inspection. While ensuring accuracy, it also takes into account production efficiency. The detection device, equipment and medium can also achieve corresponding technical effects by executing the above method instructions. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of a method for detecting abnormality in a battery formation process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0025] Embodiment 1:

[0026] like Figure 1 As shown, an embodiment of the present invention provides a method for detecting abnormality in a battery formation process, comprising the following steps: Step S1, obtaining formation parameters of each battery in the battery formation process, and preprocessing the formation parameters.

[0027] The formation parameters include more specific data. Specifically in this embodiment, the formation voltage and formation capacity are comprehensively selected according to factors such as abnormality of the battery itself, blockage of the nozzle of the formation cabinet, or abnormality of the detection equipment, so as to obtain the required abnormal detection results, avoid selecting too much data, resulting in large calculation amount and slow detection efficiency. In other optional embodiments, the technician can select appropriate formation parameters according to actual needs.

[0028] Specifically in this embodiment, preprocessing the formation parameters includes removing duplicates, removing blanks and standardizing the formation parameters.

[0029] The above-mentioned preprocessing process helps to improve data quality and provide a reliable basis for subsequent data analysis and mining. Data deduplication refers to identifying and deleting duplicate data records from a data set to ensure the uniqueness of the data. Data de-nulling refers to deleting null values ​​or missing values ​​in a data set. Null values ​​may be caused by incomplete data entry, data loss, etc. The existence of null values ​​will affect the accuracy of data analysis and mining, so it is necessary to process them. Data standardization refers to the process of converting data of different formats, structures or meanings into a unified standard format. As in the present embodiment, the waveform data of the formation voltage and the formation capacity are stored in CSV or xlsx format. The time information is converted into datetime type. Data standardization helps to improve the comparability and analyzability of data. In other optional embodiments, technicians can select other data processing means to perform data preprocessing as needed.

[0030] Step S2, calculating the mean value of the formation parameters of each section in the battery formation process according to the pre-processed formation parameters.

[0031] Specifically in this embodiment, there are 9 sections in the chemical formation process, and the mean is calculated for each section, and a 1×9-dimensional vector will be obtained for the nine sections.

[0032] Step S3: using the isolation forest algorithm to perform anomaly detection based on the mean values ​​of the formation parameters of each process section.

[0033] Isolation Forest (iForest) is a machine learning algorithm for anomaly detection. It is based on a tree structure and can quickly segment data. It does not require labeled training data and is suitable for unsupervised learning scenarios. It has a linear time complexity, which makes it very efficient when processing large-scale data. It has high accuracy in anomaly detection and can accurately identify abnormal data points.

[0034] The specific anomaly detection process includes: Step S3.1, using the isolated Senli algorithm to divide the mean value of the formation parameters of each section corresponding to each battery, and counting the number of divisions corresponding to each battery; Step S3.2: The battery whose number of splitting times is greater than the number threshold is determined to be abnormal, and the battery whose number of splitting times is less than or equal to the number threshold is determined to be normal.

[0035] Among them, the isolated Senli algorithm is used to divide the mean value of the formation parameters of each section corresponding to each battery, including: Step (1), take 0 as the base point, and repeat the following steps (2) to (3) until no split point can be selected: Step (2), calculating the formation parameter measurement of each battery according to the mean value of the formation parameter of each process section, and determining the measurement range according to the maximum and minimum values ​​of the formation parameter measurement; Step (3), randomly select a formation parameter of a battery as a split point, and at the same time satisfy that the formation parameter measurement of the selected battery is within the measurement range, and use the split point as the base point.

[0036] The parameterized metric is expressed by Euclidean distance; When 0 is taken as the base point, the expression is: ; In the formula, When 0 is the base point The formation parameter measurement of a battery, For the The battery corresponding to The mean value of the formation parameters of each process section, , is the number of work sections; When the split point is used as the base point, the expression is: ; In the formula, When the split point is used as the base point The formation parameter measurement of a battery, The first The mean value of the formation parameters of each process section.

[0037] After obtaining the abnormality judgment of each battery, make a comprehensive abnormality judgment: A. For a single battery, if both the formation voltage and the formation capacity are judged to be abnormal, the battery is judged to have a formation abnormality; if the number of batteries with formation abnormalities on the same tray exceeds the quantity threshold, the nozzle corresponding to the tray is judged to be abnormal; B. For a single battery, if the formation voltage is judged to be abnormal, and the formation capacity is judged to be normal, the battery is judged to have a formation voltage abnormality; if the number of batteries with formation voltage abnormalities on the same tray exceeds the quantity threshold, the voltage detection device corresponding to the tray is judged to be abnormal; C. For a single battery, if no abnormality is found based on the formation voltage and an abnormality is found based on the formation capacity, the battery is judged to have an abnormal formation capacity; if the number of batteries with abnormal formation capacity on the same tray exceeds the quantity threshold, the capacity detection device corresponding to the tray is judged to be abnormal; D. For a single battery, if both the formation voltage and the formation capacity are judged to be normal, then it is determined that the battery has no formation abnormality.

[0038] In the above scheme, the single battery and the batteries on the entire tray are analyzed separately according to the formation voltage and formation capacity, and the abnormality judgment is made on the battery itself, the voltage detection equipment, the capacity detection equipment, and the nozzle of the formation cabinet respectively, which reduces the production waste of manual inspection due to shutdown and has excellent detection effect.

[0039] In summary, the battery formation process anomaly detection method proposed in this embodiment improves the reliability of the formation parameters by collecting, preprocessing and averaging the formation parameters, and ensures the accuracy of subsequent anomaly detection processing. The processed formation parameters are detected by using the isolation forest algorithm. Compared with the traditional scheme, it is not affected by human subjective experience and does not require shutdown inspection. While ensuring accuracy, it also takes into account production efficiency.

[0040] Embodiment 2:

[0041] An embodiment of the present invention provides a battery formation process abnormality detection device, comprising: A parameter acquisition processing module is configured to acquire the formation parameters of each battery in the battery formation process and pre-process the formation parameters; A formation parameter calculation module is configured to calculate the mean value of the formation parameters of each section in the battery formation process according to the pre-processed formation parameters; The anomaly detection and judgment module is configured to use the isolation forest algorithm to perform anomaly detection based on the mean value of the formation parameters of each process section.

[0042] Embodiment three:

[0043] Based on the battery formation process abnormality detection method provided in the first embodiment, the embodiment of the present invention provides an electronic device, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0044] Embodiment 4:

[0045] Based on the battery formation process abnormality detection method provided in Example 1, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented.

[0046] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0047] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0048] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0049] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0050] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for detecting abnormality in a battery formation process, characterized in that: include: Acquiring formation parameters of each battery during the battery formation process, and preprocessing the formation parameters; Calculate the mean value of the formation parameters of each section in the battery formation process according to the pre-treated formation parameters; The isolation forest algorithm is used to detect anomalies based on the mean values ​​of the formation parameters of each process section.

2. The battery formation process abnormality detection method according to claim 1, characterized in that: The formation parameters include formation voltage and formation capacity.

3. The battery formation process abnormality detection method according to claim 1, characterized in that: The preprocessing of the formation parameters includes deduplication, removal of blanks and standardization of the formation parameters.

4. The battery formation process abnormality detection method according to claim 1, characterized in that: The method of using an isolation forest algorithm to perform anomaly detection based on the mean value of the formation parameters of each process section includes: The isolated Senli algorithm is used to divide the mean value of the formation parameter of each section corresponding to each battery, and the number of divisions corresponding to each battery is counted; A battery whose number of cuts is greater than a number threshold is determined to be abnormal, and a battery whose number of cuts is less than or equal to the number threshold is determined to be not abnormal.

5. The battery formation process abnormality detection method according to claim 4, characterized in that: The method of using the isolated Senli algorithm to divide the mean value of the formation parameter of each section corresponding to each battery includes: Use 0 as the base point and repeat the following steps until no more split points can be selected: Calculating the formation parameter metric of each battery according to the formation parameter mean of each section, and determining the metric range according to the maximum and minimum values ​​of the formation parameter metric; A formation parameter of a battery is randomly selected as a split point, and the formation parameter measurement of the selected battery is within the measurement range, and the split point is used as a base point.

6. The battery formation process abnormality detection method according to claim 5, characterized in that: The transformation parameter metric is represented by Euclidean distance; When 0 is taken as the base point, the expression is: ； In the formula, When 0 is the base point The formation parameter measurement of a battery, For the The battery corresponding to The mean value of the formation parameters of each process section, , is the number of work sections; When the split point is used as the base point, the expression is: ； In the formula, When the split point is used as the base point The formation parameter measurement of a battery, The first The mean value of the formation parameters of each process section.

7. The battery formation process abnormality detection method according to claim 2, characterized in that: For a single battery, if both the formation voltage and the formation capacity are judged to be abnormal, then the battery is determined to have a formation abnormality; If the number of batteries with abnormal formation on the same tray exceeds the quantity threshold, it is determined that the nozzle corresponding to the tray is abnormal; For a single battery, if it is determined that there is an abnormality based on the formation voltage, and it is determined that there is no abnormality based on the formation capacity, then it is determined that the battery has a formation voltage abnormality; If the number of batteries with abnormal formation voltage on the same tray exceeds the quantity threshold, it is determined that the voltage detection device corresponding to the tray is abnormal; For a single battery, if it is determined that there is no abnormality based on the formation voltage, and that there is an abnormality based on the formation capacity, then it is determined that the battery has a formation capacity abnormality; If the number of batteries with abnormal formation capacity on the same tray exceeds the quantity threshold, it is determined that the capacity detection device corresponding to the tray is abnormal; For a single battery, if it is determined that there is no abnormality based on both the formation voltage and the formation capacity, it is determined that there is no formation abnormality in the battery.

8. A battery formation process abnormality detection device, characterized in that: include: A parameter acquisition processing module is configured to acquire the formation parameters of each battery in the battery formation process and pre-process the formation parameters; A formation parameter calculation module is configured to calculate the mean value of the formation parameters of each section in the battery formation process according to the pre-processed formation parameters; The abnormality detection and judgment module is configured to perform abnormality detection using an isolation forest algorithm based on the mean value of the formation parameters of each process section.

9. An electronic device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images