Production data anomaly detection method and device, equipment, storage medium and product

Abstract

The embodiment of the invention provides a production data anomaly detection method and device, equipment, a storage medium and a product, and the method comprises the steps: obtaining a production curve corresponding to the production data of a to-be-detected product, and enabling the horizontal axis of the production curve to be used for indicating a set of production specification parameter values, the longitudinal axis of the production curve is used for indicating a set of frequencies corresponding to each production specification parameter value; counting the number of wave crests on the production curve; and performing anomaly detection based on the number of wave crests. Therefore, whether production is abnormal or not can be judged based on whether the number of the wave crests changes or not, the production condition can be monitored more comprehensively, and the accuracy of abnormal detection of the production data is improved.

Description

Technical Field

[0001] The present application relates to the field of production and manufacturing technology, and in particular to a method, device, equipment, storage medium and product for detecting anomalies in production data. Background Art

[0002] In industrial production, ensuring product quality and improving production efficiency are the core goals pursued by enterprises. However, due to the diversity of equipment performance and production environment, including fluctuations in process parameters, differences in operator skills, inconsistency in raw materials, and wear or aging of equipment, it is difficult for the products produced to meet completely consistent standards. Therefore, it is necessary to perform anomaly detection on production data to detect abnormal products, equipment or production lines.

[0003] At present, anomaly detection methods usually use statistical indicators such as the mean and standard deviation of production data for anomaly detection. Such indicator data can only describe the central trend and dispersion of production data, but are insensitive to slight changes in the distribution of production data, resulting in poor accuracy of production data anomaly detection. Summary of the invention

[0004] The embodiments of the present application provide a production data anomaly detection method, device, equipment, storage medium and product to solve the technical problem of poor accuracy in production data anomaly detection.

[0005] In a first aspect, an embodiment of the present application provides a method for detecting anomalies in production data, comprising: Obtaining a production curve corresponding to the production data of the product to be tested, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value; Count the number of peaks on the production curve; Anomaly detection is performed based on the number of peaks.

[0006] In this embodiment, a production curve corresponding to the production data of the product to be tested can be obtained. The production curve can have a set of production specification parameter values ​​as the horizontal axis and a set of frequencies corresponding to each production specification parameter value as the vertical axis. The frequency of occurrence of the same production specification parameter value can be reflected, thereby counting the number of peaks on the production curve. Generally speaking, in normal production processes, the number of peaks is relatively stable, and slight changes in the distribution of production data can be reflected in the production curve, and the number of peaks will also change accordingly. Therefore, whether the production is abnormal can be judged based on whether the number of peaks changes, which can more comprehensively monitor the production situation and improve the accuracy of production data anomaly detection.

[0007] In some embodiments, anomaly detection is performed based on the number of peaks, including: When the number of peaks is greater than a first threshold, it is determined that the production data of the product to be detected is abnormal, and the first threshold is a critical value of the number of peaks that characterizes the abnormal production data.

[0008] In this way, the number of peaks of the production curve corresponding to the production data of a certain production line or a certain device is counted, and based on the set first threshold, it is determined whether the production data of the production line or the device is abnormal, thereby realizing single-point dimensional production data anomaly detection.

[0009] In some embodiments, the production curve includes production curves corresponding to multiple production cycles, and abnormality detection is performed based on the number of peaks, including: Compare the number of peaks on the production curve corresponding to multiple production cycles; When the number of peaks on the production curves corresponding to the plurality of production cycles changes, it is determined that the production data of the product to be detected is abnormal.

[0010] In this way, for the production curves corresponding to multiple production cycles, the number of peaks in each production cycle can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production cycles shows a significant change, it can be considered that the distribution of production data in multiple production cycles is abnormal, thereby realizing production data anomaly detection in the time dimension.

[0011] In some embodiments, the production curve includes production curves corresponding to multiple production lines, and abnormality detection is performed based on the number of peaks, including: Compare the number of peaks on the production curves corresponding to multiple production lines; When the numbers of peaks on the production curves corresponding to the multiple production lines are not equal, it is determined that the production data of the product to be detected is abnormal.

[0012] In this way, for the production curves corresponding to multiple production lines, the number of peaks of each production line can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production lines is unequal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0013] In some embodiments, the production curve includes production curves corresponding to multiple devices on the same production line, and anomaly detection is performed based on the number of peaks, including: Compare the number of peaks on the production curves corresponding to multiple devices; When the numbers of peaks on the production curves corresponding to the plurality of devices are not equal, it is determined that the production data of the product to be detected is abnormal.

[0014] In this way, for the production curves corresponding to multiple devices on the same production line, the number of peaks of each device can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple devices is not equal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0015] In some embodiments, before counting the number of peaks on the production curve, the method further comprises: Determine the peak on the production curve, the ordinate of the peak is greater than the ordinates of other points in the neighborhood; wherein the distance between the abscissas of two adjacent peaks is greater than or equal to the distance threshold.

[0016] In this way, there will be only one peak in the same neighborhood, and there is a distance limit between two adjacent peaks, which can ensure the independence of each peak.

[0017] In some embodiments, determining a peak in a production curve includes: Determine the initial point on the production curve, the ordinate of the initial point is greater than the ordinates of other points in the neighborhood; When the ordinate of the preliminary selected point is greater than or equal to the height threshold, the preliminary selected point is determined as the peak on the production curve.

[0018] Thus, in order to further improve the accuracy and relevance of peak identification, a height threshold is introduced. Only when the ordinate exceeds the height threshold, the corresponding preliminary point is officially identified as a peak.

[0019] In some embodiments, obtaining a production curve corresponding to production data of the product to be tested includes: Obtain production data of the product to be tested; Smoothing of production data; Generate a production curve based on the smoothed production data.

[0020] In this way, the production data can be smoothed to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0021] In some embodiments, the production data is smoothed, including: For each data point in the production data, a neighborhood of the data point is determined based on a ratio of the number of the data points; Assigning weights to all data points in the neighborhood, wherein the weight of the first data point is greater than the weight of the second data point, and the distance between the first data point and the center point in the neighborhood is less than the distance between the second data point and the center point; A local weighted linear regression operation is performed according to the weights of all data points in the neighborhood to obtain the smoothed coordinates of the data points.

[0022] In this way, the neighborhood size can be determined based on the parameter value, and the production data can be smoothed based on the local weighted linear regression operation to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0023] In some embodiments, a local weighted linear regression operation is performed according to the weights of all data points in the neighborhood to obtain smoothed coordinates of the data points, including: Construct a weighted matrix based on the weights of all data points in the neighborhood; Determine a local weighted linear regression equation according to the weighted matrix and the linear regression equation; Solve the model parameters in the locally weighted linear regression equation; Determine the smoothed coordinates of the data points based on the model parameters and the coordinates of the data points.

[0024] In this way, a local weighted linear regression equation can be constructed. After solving the model parameters, the smooth coordinates are calculated based on the model parameters and the coordinates of the data points to achieve the purpose of smoothing the data and improving the data quality.

[0025] In some embodiments, after performing a local weighted linear regression operation according to the weights of all data points in the neighborhood to obtain smoothed coordinates of the data points, the method further includes: Generate an initial curve based on the smoothed production data; Count the number of peaks on the initial curve; In the case where the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio; Smoothing of production data based on adjusted quantity ratios; When the number of peaks on the initial curve does not exceed a standard threshold range, the initial curve is determined as a production curve; The maximum and minimum values ​​of the standard threshold range are critical values ​​of the number of peaks that characterize the noise of the production data and meet the smoothing requirement.

[0026] In this way, the initial curve generated by the smoothed production data can be checked to determine whether the initial curve meets the requirements. If the number of peaks on the initial curve exceeds the standard threshold range, it can be considered that the initial curve does not meet the requirements. At this time, the number ratio can be adjusted to redetermine the neighborhood size and then smoothed, which further improves the accuracy of the production curve and thereby improves the accuracy of production data anomaly detection. If the number of peaks on the initial curve does not exceed the standard threshold range, it can be considered that the initial curve is a production curve that meets the requirements. At this time, accurate production data anomaly detection results can be obtained based on the production curve.

[0027] In some embodiments, when the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio includes: In the case where the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, the number ratio is increased.

[0028] Thus, if the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, it can be considered that the noise of the initial curve is still high. In this case, the number ratio can be increased and smoothing can be performed again to further reduce the noise.

[0029] In some embodiments, when the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio includes: In the case where the number of peaks on the initial curve is smaller than the minimum value of the standard threshold range, the number ratio is reduced.

[0030] Thus, if the number of peaks on the initial curve is less than the minimum value of the standard threshold range, it can be considered that the initial curve is over-smoothed. In this case, the number ratio can be reduced and smoothing can be performed again to obtain a more accurate production curve.

[0031] In a second aspect, an embodiment of the present application further provides a production data anomaly detection device, comprising: An acquisition module, used to acquire a production curve corresponding to the production data of the product to be detected, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value; A statistical module, used for counting the number of peaks on the production curve; A detection module is used to perform anomaly detection based on the number of peaks.

[0032] In this embodiment, a production curve corresponding to the production data of the product to be tested can be obtained. The production curve can have a set of production specification parameter values ​​as the horizontal axis and a set of frequencies corresponding to each production specification parameter value as the vertical axis. The frequency of occurrence of the same production specification parameter value can be reflected, thereby counting the number of peaks on the production curve. Generally speaking, in normal production processes, the number of peaks is relatively stable, and slight changes in the distribution of production data can be reflected in the production curve, and the number of peaks will also change accordingly. Therefore, whether the production is abnormal can be judged based on whether the number of peaks changes, which can more comprehensively monitor the production situation and improve the accuracy of production data anomaly detection.

[0033] In a third aspect, an embodiment of the present application provides a battery production line, which uses the production data anomaly detection method of the first aspect to perform anomaly detection on production data.

[0034] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory storing program instructions; when the processor executes the program instructions, the method of the first aspect is implemented.

[0035] In a fourth aspect, an embodiment of the present application provides a machine-readable storage medium having program instructions stored thereon, and the method of the first aspect is implemented when the program instructions are executed by a processor.

[0036] In a fifth aspect, an embodiment of the present application provides a computer program product. When instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method of the first aspect.

[0037] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the drawings without paying creative work.

[0039] Figure 1 A flowchart of a production data anomaly detection method provided in an embodiment of the present application; Figure 2 A schematic diagram of single-point dimension anomaly detection in the production data anomaly detection method provided in an embodiment of the present application; Figure 3 A schematic diagram of time dimension anomaly detection in the production data anomaly detection method provided in an embodiment of the present application; Figure 4 One of the schematic diagrams of spatial dimension anomaly detection in the production data anomaly detection method provided in an embodiment of the present application; Figure 5 The second schematic diagram of spatial dimension anomaly detection in the production data anomaly detection method provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of a production data anomaly detection device provided in an embodiment of the present application; Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0040] In the drawings, the drawings are not drawn to scale. DETAILED DESCRIPTION

[0041] The following detailed description and drawings of the embodiments of the present application are used to illustrate the principles of the present application, but cannot be used to limit the scope of the present application, that is, the present application is not limited to the described embodiments.

[0042] In the description of the present application, it should be noted that, unless otherwise specified, "multiple" means more than two; the terms "upper", "lower", "left", "right", "inside", "outside", etc., indicating the orientation or positional relationship, are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. "Vertical" is not strictly vertical, but is within the allowable error range. "Parallel" is not strictly parallel, but is within the allowable error range.

[0043] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0044] The embodiments of the present application provide a production data anomaly detection method, device, equipment, storage medium and product to solve the above technical problems. The production data anomaly detection method provided by the embodiments of the present application is first introduced below.

[0045] Please refer to Figure 1 , Figure 1 A flowchart of a production data anomaly detection method provided in an embodiment of the present application is provided. The production data anomaly detection method may include: Step 101, obtaining a production curve corresponding to production data of a product to be tested, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value.

[0046] In step 101, the production specification parameter value can represent the numerical value of the production specification of the product to be tested, where the production specifications may include size specifications, performance specifications, etc. In order to facilitate understanding of the technical solutions of the embodiments of the present application, the following will take the product to be tested as a battery and the production specification as the thickness of the battery cell as an example. The production specification parameter value can be the measured value of the thickness of the battery cell.

[0047] In this embodiment, production data of the product to be tested during a certain period of time (which may include one or more production cycles), each production line or each device during the production process can be collected. The production data may include multiple data points (x, y). Among them, x can be a production specification parameter value. Taking the product to be tested as a battery as an example, the production specification parameter value can be a measured value of the thickness of the battery cell. y can be the frequency of each production specification parameter value. For example, the measured value of the thickness of the battery cell during the production process is that the frequency of thickness 1 is A times, then a data point in the production data can be expressed as (thickness 1, A).

[0048] It is understandable that the production data of multiple production cycles may refer to the production data corresponding to one or more production lines that produce the product to be detected in different production cycles. The production data of multiple production lines may refer to the production data corresponding to the production of the product to be detected by different production lines in one production cycle or multiple production cycles. The production data of multiple devices may refer to the production data corresponding to the production of the product to be detected by different devices of the same production line in one production cycle or multiple production cycles. In order to facilitate the understanding of the technical solution provided in the embodiment of the present application, the production data of multiple production cycles will be used as the production data corresponding to all production lines that produce the product to be detected in different production cycles, the production data of multiple production lines will be the production data corresponding to the production of the product to be detected by different production lines in one production cycle, and the production data of multiple devices will be the production data corresponding to the production of the product to be detected by different devices of the same production line in one production cycle.

[0049] Among them, taking the production cycle of one day as an example, the set of production data of multiple production cycles can be expressed as: {'day1': 'x':[…],'y':[…],'day2': 'x':[…],'y':[…]…}. The set of production data of multiple production lines can be expressed as: {line1:{'x':[…],'y':[…]},line2:{'x':[…],'y':[…]}…}. And the set of production data of multiple devices on the same production line can be expressed as: {m1:{'x':[…],'y':[…]…},m2:{'x':[…],'y':[…]}…}. Where x and y have the same length.

[0050] A corresponding production curve may be generated based on the production data. The horizontal axis of the production curve may be used to indicate a set of production specification parameter values, and the vertical axis of the production curve may be used to indicate a set of frequencies corresponding to each production specification parameter value.

[0051] Step 102: Count the number of peaks on the production curve.

[0052] In step 102, peaks may be determined from the production curve, and the number of peaks may be counted.

[0053] Step 103: perform anomaly detection based on the number of peaks.

[0054] It is understandable that the phenomenon of multi-peak distribution of production curves corresponding to production data in industrial production can usually be attributed to the following reasons: 1. Process changes or failures: Changes in process parameters or equipment failures may occur during the production process, which may lead to changes in product quality or performance, and thus affect the distribution characteristics of production data. For example, changes in process parameters such as material composition, temperature, and pressure may lead to changes in product characteristics, which are reflected in the distribution of production data.

[0055] 2. Inconsistent human operation: The operator's technical level or operating habits may affect the distribution of production data. Different operators may produce different product characteristics or quality under the same production conditions, resulting in changes in the distribution of production data.

[0056] 3. Raw material differences: If the raw materials used have batch differences or come from different suppliers, these differences may result in different performance or quality of the final product, which in turn affects the distribution characteristics of the production data.

[0057] 4. Changes in environmental factors: Changes in environmental factors inside and outside the factory, such as changes in temperature, humidity and other conditions, may affect the production process, thereby changing the characteristics of the product and the corresponding production data distribution.

[0058] 5. Equipment wear or aging: Long-term use of production equipment may cause changes or wear in its performance. These changes may indirectly affect product quality and thus affect the distribution of production data.

[0059] 6. Data recording or sampling problems: Problems in the data recording or sampling process (such as insufficient sampling frequency, accuracy, etc.) may cause the observed production data distribution to be inconsistent with expectations, resulting in offset, tailing, or multi-peak phenomena.

[0060] Therefore, there are many reasons for abnormal distribution of production data in industrial production. It is usually necessary to determine the specific reasons through detailed data analysis and production process monitoring, and take corresponding measures to adjust or improve to ensure the stability of product quality and production efficiency. Based on this, the number of peaks can be used as the basis for determining whether the production data is abnormal.

[0061] In step 103, the number of peaks may be compared with a preset threshold value, and an abnormality detection result may be obtained based on the comparison result. Alternatively, the number of peaks on the production curves corresponding to different production lines or different devices on the same production line may be compared, and an abnormality detection result may be obtained based on the comparison result. Alternatively, the number of peaks on the production curves corresponding to different production cycles may be compared, and an abnormality detection result may be obtained based on the comparison result.

[0062] In this embodiment, a production curve corresponding to the production data of the product to be tested can be obtained. The production curve can have a set of production specification parameter values ​​as the horizontal axis and a set of frequencies corresponding to each production specification parameter value as the vertical axis. The frequency of occurrence of the same production specification parameter value can be reflected, thereby counting the number of peaks on the production curve. Generally speaking, in normal production processes, the number of peaks is relatively stable, and slight changes in the distribution of production data can be reflected in the production curve, and the number of peaks will also change accordingly. Therefore, whether the production is abnormal can be judged based on whether the number of peaks changes, which can more comprehensively monitor the production situation and improve the accuracy of production data anomaly detection.

[0063] In some embodiments, anomaly detection is performed based on the number of peaks, including: When the number of peaks is greater than a first threshold, it is determined that the production data of the product to be detected is abnormal, and the first threshold is a critical value of the number of peaks that characterizes the abnormal production data.

[0064] In this embodiment, for the production data within a production cycle of a production line, or the integrated production data within multiple production cycles, the number of peaks of the corresponding production curve can be compared with a first threshold set in advance to determine whether the production data of the product to be tested produced by the production line is abnormal.

[0065] It is understandable that the first threshold may be a critical value of the number of peaks that characterize abnormal production data, that is, if the first threshold is exceeded, the production data may be considered abnormal. The first threshold may be set according to actual conditions, and different products to be tested may have different corresponding first thresholds.

[0066] When the number of peaks is greater than the first threshold, it can be considered that the fluctuation of the production data is large, that is, it can be determined that the production data of the product to be detected is abnormal.

[0067] For example, Figure 2 As shown, anomaly detection may include the following steps: Step 201, counting the number of peaks of the production curve corresponding to the production data of the product to be tested in a certain time period and on a certain production line; Step 202, determining whether the number of peaks is greater than a first threshold; if so, executing step 203; if not, executing step 204; Step 203: the production data of the product to be tested in the production line during the time period is abnormal; Step 204: the production data of the product to be tested in this time period and on this production line is normal.

[0068] In this way, the number of peaks of the production curve corresponding to the production data of a certain production line or a certain device is counted, and based on the set first threshold, it is determined whether the production data of the production line or the device is abnormal, thereby realizing single-point dimensional production data anomaly detection.

[0069] In some embodiments, the production curve includes production curves corresponding to multiple production cycles, and abnormality detection is performed based on the number of peaks, including: Compare the number of peaks on the production curve corresponding to multiple production cycles; When the number of peaks on the production curves corresponding to the plurality of production cycles changes, it is determined that the production data of the product to be detected is abnormal.

[0070] In this embodiment, for the production data of multiple production cycles, the number of peaks of the production curve corresponding to each production cycle can be counted and compared. It can be understood that in the normal production process, the production data of each production cycle is relatively stable, so the number of peaks on the production curve corresponding to each production cycle is roughly the same. In the case where the number of peaks on the production curve corresponding to multiple production cycles changes, for example, the number of peaks increases significantly, it can be considered that the distribution of production data is abnormal within the time period of multiple production cycles, and it can be determined that the production data of the product to be detected is abnormal.

[0071] For example, Figure 3 As shown, anomaly detection may include the following steps: Step 301, counting the number of peaks of the production curve corresponding to the production data of the product to be tested in multiple production cycles; Step 302, determine whether the number of peaks has changed; if so, execute step 303; if not, execute step 304; Step 303, the production data of the product to be detected is abnormal; Step 304: the production data of the product to be tested is normal.

[0072] In this way, for the production curves corresponding to multiple production cycles, the number of peaks in each production cycle can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production cycles shows a significant change, it can be considered that the distribution of production data in multiple production cycles is abnormal, thereby realizing production data anomaly detection in the time dimension.

[0073] In some embodiments, the production curve includes production curves corresponding to multiple production lines, and abnormality detection is performed based on the number of peaks, including: Compare the number of peaks on the production curves corresponding to multiple production lines; When the numbers of peaks on the production curves corresponding to the multiple production lines are not equal, it is determined that the production data of the product to be detected is abnormal.

[0074] In this embodiment, for the production data under multiple production lines within one production cycle or multiple production cycles, the number of peaks of the production curve corresponding to each production line can be counted and compared. It can be understood that in a normal production process, the production data between different production lines should be consistent, so the number of peaks on the production curves corresponding to multiple production lines is equal. In the case where the number of peaks on the production curves corresponding to multiple production lines is not equal, it can be considered that the production data of the product to be tested under different production lines is inconsistent, and it can be determined that the production data of the product to be tested is abnormal.

[0075] For example, Figure 4 As shown, anomaly detection may include the following steps: Step 401, counting the number of peaks of the production curve corresponding to the production data of the product to be tested under multiple production lines; Step 402, determine whether the number of peaks is equal; if not, execute step 403; if equal, execute step 404; Step 403, the production data of the product to be tested has poor consistency; Step 404: The production data of the product to be tested has good consistency.

[0076] In this embodiment, if the number of peaks on the production curve corresponding to a certain production line is found to be significantly different from other production lines, this may indicate that the production line has an abnormal condition or a specific process state at the same time, which requires further investigation and analysis. The production line code, abnormal type, abnormal value, and date output can be saved as abnormal information. The system can automatically feed back the abnormal information to the production link, prompting the operator to make necessary adjustments, including equipment maintenance, process parameter adjustment, or strengthening of product quality control measures, to ensure the yield rate of the product to be tested.

[0077] In this way, for the production curves corresponding to multiple production lines, the number of peaks of each production line can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production lines is unequal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0078] In some embodiments, the production curve includes production curves corresponding to multiple devices on the same production line, and anomaly detection is performed based on the number of peaks, including: Compare the number of peaks on the production curves corresponding to multiple devices; When the numbers of peaks on the production curves corresponding to the plurality of devices are not equal, it is determined that the production data of the product to be detected is abnormal.

[0079] In this embodiment, for the production data of multiple devices on the same production line within one production cycle or multiple production cycles, the number of peaks of the production curve corresponding to each device can be counted and compared. It can be understood that in a normal production process, the production data between different devices on the same production line should be consistent, so the number of peaks on the production curve corresponding to multiple devices on the same production line is equal. In the case where the number of peaks on the production curve corresponding to multiple devices is not equal, it can be considered that the production data of different devices on the same production line of the product to be tested is inconsistent, and it can be determined that the production data of the product to be tested is abnormal.

[0080] For example, Figure 5 As shown, anomaly detection may include the following steps: Step 501, counting the number of peaks of the production curve corresponding to the production data of multiple devices on the same production line of the product to be tested; Step 502, determine whether the number of peaks is equal; if not, execute step 503; if equal, execute step 504; Step 503, the production data of the product to be tested has poor consistency; Step 504: The production data of the product to be tested has good consistency.

[0081] In this embodiment, if the number of peaks on the production curve corresponding to a certain device is found to be significantly deviated from other devices on the same production line, this may indicate that the device is in an abnormal condition or a specific process state at the same time, and further investigation and analysis is required. The production line code, equipment code, abnormal type, abnormal value, and date output can be saved as abnormal information. The system can automatically feed back abnormal information to the production link, prompting the operator to make necessary adjustments, including equipment maintenance, process parameter adjustment, or strengthening of product quality control measures, etc., to ensure the yield rate of the product to be tested.

[0082] In this way, for the production curves corresponding to multiple devices on the same production line, the number of peaks of each device can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple devices is not equal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0083] In some embodiments, before counting the number of peaks on the production curve, the method further comprises: The peak on the production curve is determined, and the ordinate of the peak is greater than the ordinates of other points in the neighborhood; wherein the distance between the abscissas of two adjacent peaks is greater than or equal to the distance threshold.

[0084] In this embodiment, the neighborhood can represent the surrounding area of ​​any point. The δ neighborhood of point x0 is the set of all points that satisfy the inequality |x−x0|<δ, denoted as U δ (x0)=x||x−x0|<δ.

[0085] The size of the neighborhood can be determined based on a preset ratio of the number of data points. The value of the ratio can be related to the product to be inspected. For products to be inspected that require high production accuracy, the corresponding ratio is relatively small. For example, if the ratio frac=0.1, the 10% of the nearest points can be included in the same neighborhood.

[0086] For the same neighborhood, a local maximum value on the production curve, i.e., a peak value, can be identified. It can be understood that the peak value can be the ordinate of the peak, in other words, the ordinate of the peak is greater than the ordinates of other points in the neighborhood.

[0087] For example, the ordinates of all points in the neighborhood can be obtained, and the point with the largest ordinate is determined as the peak. In some examples, the peak detection can be performed on the ordinate of the production curve directly with the help of a peak finding function (find_peaks function) in an open source Python (Scientific Python, SciPy) library. This function can identify the local maximum point by comparing the values ​​of the ordinates, and then determine the peak.

[0088] In order to prevent the peak recognition from being too dense, in this embodiment, the minimum distance between the horizontal coordinates of two adjacent peaks can also be controlled by setting a distance threshold. The distance threshold can be set according to the empirical value in combination with the actual situation. For example, the distance threshold can be 0.2mm, which means that there is only one peak in the measured value of the battery cell thickness within 0.2mm. If there are two peaks with horizontal coordinates less than 0.2mm, the point with the larger vertical coordinate is determined as the peak to be counted, and the other point is screened out and no peak statistics are performed.

[0089] In this way, there will be only one peak in the same neighborhood, and there is a distance limit between two adjacent peaks, which can ensure the independence of each peak.

[0090] In some embodiments, determining a peak in a production curve includes: Determine the initial point on the production curve, the ordinate of the initial point is greater than the ordinates of other points in the neighborhood; When the ordinate of the preliminary selected point is greater than or equal to the height threshold, the preliminary selected point is determined as the peak on the production curve.

[0091] In this embodiment, the height threshold is the value of the minimum ordinate that characterizes the peak. A preliminary point with a ordinate that is a local maximum value can be determined from the production curve, and the ordinate of the preliminary point can be compared with the height threshold. When the ordinate of the preliminary point is greater than or equal to the height threshold, the preliminary point is determined as the peak on the production curve. If it is less than the height threshold, it means that the preliminary point is not a peak, and when counting the number of peaks, the preliminary point is not counted. This helps to filter out false positive peaks caused by small fluctuations or noise, and ensure that the identified peaks truly reflect the significant features in the production data.

[0092] Thus, in order to further improve the accuracy and relevance of peak identification, a height threshold is introduced. Only when the ordinate exceeds the height threshold, the corresponding preliminary point is officially identified as a peak.

[0093] In some embodiments, obtaining a production curve corresponding to production data of the product to be tested includes: Obtain production data of the product to be tested; Smoothing of production data; Generate a production curve based on the smoothed production data.

[0094] In this embodiment, the collected production data may be smoothed to reduce noise and improve data quality. For example, the production data may be smoothed by simple moving average (SMA), weighted moving average (WMA), exponential smoothing, Gaussian smoothing, and Savitzky-Golay filter smoothing, which are not specifically limited here.

[0095] A production curve can be generated based on the smoothed production data.

[0096] In this way, the production data can be smoothed to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0097] In some embodiments, the production data is smoothed, including: For each data point in the production data, determine a neighborhood size of the data point based on a ratio of the number of data points; Assigning weights to all data points in the neighborhood, wherein the weight of the first data point is greater than the weight of the second data point, and the distance between the first data point and the center point in the neighborhood is less than the distance between the second data point and the center point; A local weighted linear regression operation is performed according to the weights of all data points in the neighborhood to obtain the smoothed coordinates of the data points.

[0098] In this implementation, a data point (x, y) can be selected as the center point, and the neighborhood size is determined based on the number ratio of data points frac. The number ratio frac is the fraction of data points used to determine the number of data points used in the local regression. For example, if frac=0.1, the nearest 10% of data points are used for local regression.

[0099] Weights can be assigned to all data points in the neighborhood, and a triple truncated Gaussian kernel function can be used to assign weights, where the weight of the first data point is greater than the weight of the second data point, and the distance between the first data point and the center point in the neighborhood is less than the distance between the second data point and the center point. In other words, the closer the data point is to the center point, the greater the weight.

[0100] Perform a local weighted linear regression operation based on the weights of all data points in the neighborhood to obtain the smoothed coordinates of the center point (x_smooth, y_smooth). Repeat the above steps for each data point in the production data to obtain the smoothed production data.

[0101] In this way, the neighborhood size can be determined based on the parameter value, and the production data can be smoothed based on the local weighted linear regression operation to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0102] In some embodiments, performing a local weighted linear regression operation according to the weights of all data points in the neighborhood to obtain the smoothed coordinates of the data points may include: Construct a weighted matrix based on the weights of all data points in the neighborhood; According to the weighted matrix and the linear regression equation, a local weighted linear regression equation is determined; Solve the model parameters in the locally weighted linear regression equation; Determine the smoothed coordinates of the data points based on the model parameters and the coordinates of the data points.

[0103] In this embodiment, local weighted linear regression is a non-parametric algorithm that can calculate parameters for each center point separately. When calculating parameters, points near the center point have a higher "preference" than points far from the center point. This preference can be achieved through a weight function that assigns different weights to points based on their distance from the center point.

[0104] The weighted matrix W can be constructed based on the weights of all data points in the neighborhood, where W ii =w (i) .w (i) Represents the weight of the i-th point. Introducing the weighted matrix W into the linear regression equation, we can get the weighted linear regression equation. By minimizing the weighted cost function, we can solve the model parameter θ of the local weighted linear regression. Among them, the cost function of the local weighted linear regression can be expressed as: Where m is the number of data points, θ is the model parameter, and x (i) The input value of the i-th point, i.e. the horizontal coordinate, y (i) is the output value of the i-th point, that is, the vertical coordinate.

[0105] The smooth coordinates of the center point are calculated using the model parameter θ and the coordinates of the center point. The smoothed production data can be obtained by traversing each data point in the production data.

[0106] In this way, a local weighted linear regression equation can be constructed. After solving the model parameters, the smooth coordinates are calculated based on the model parameters and the coordinates of the data points to achieve the purpose of smoothing the data and improving the data quality.

[0107] In some embodiments, after performing a local weighted linear regression operation according to the weights of all data points in the neighborhood to obtain smoothed coordinates of the data points, the method further includes: Generate an initial curve based on the smoothed production data; Count the number of peaks on the initial curve; In the case where the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio; Smoothing of production data based on adjusted quantity ratios; When the number of peaks on the initial curve does not exceed a standard threshold range, the initial curve is determined as a production curve; The maximum and minimum values ​​of the standard threshold range are critical values ​​of the number of peaks that characterize the noise of the production data and meet the smoothing requirement.

[0108] In this embodiment, the adjustment of the number ratio frac is crucial. The number ratio defines the proportion of data points used in the local regression. A smaller number ratio makes the smoothed curve closer to the original data, while a larger number ratio produces a smoother curve.

[0109] Therefore, whether the size of the number ratio is appropriate is crucial to the accuracy of the subsequent statistical results of the number of peaks. Based on this, the initial curve can be generated based on the smoothed production data. The number of peaks on the initial curve is first counted and compared with the standard threshold range to determine whether the number ratio is appropriate. When the number of peaks on the initial curve exceeds the standard threshold range, the number ratio is adjusted and the neighborhood size is re-determined before smoothing the production data to obtain a production curve that better meets the requirements. When the number of peaks on the initial curve does not exceed the standard threshold range, it can be considered that the current initial curve can clearly reflect the characteristics of the production data, and at the same time, there is less noise, which is convenient for statistical analysis. Therefore, the initial curve can be determined as the production curve.

[0110] It can be understood that the maximum and minimum values ​​of the standard threshold range are the critical values ​​of the number of peaks that characterize the noise of the production data meeting the smoothing requirements. The initial number ratio and the standard threshold range can be set based on the production accuracy requirements of the product to be tested, and are not specifically limited here.

[0111] In this way, the initial curve generated by the smoothed production data can be checked to determine whether the initial curve meets the requirements. If the number of peaks on the initial curve exceeds the standard threshold range, it can be considered that the initial curve does not meet the requirements. At this time, the number ratio can be adjusted and re-smoothing can be performed to further improve the accuracy of the production curve, thereby improving the accuracy of production data anomaly detection. If the number of peaks on the initial curve does not exceed the standard threshold range, the initial curve can be considered to be a production curve that meets the requirements. At this time, accurate production data anomaly detection results can be obtained based on the production curve.

[0112] In some embodiments, when the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio includes: In the case where the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, the number ratio is increased.

[0113] In this embodiment, when the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, for example, in the power battery production data, if a high noise level is observed based on the initial curve, the number ratio can be appropriately increased to obtain a clearer and smoother production curve.

[0114] Thus, if the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, it can be considered that the noise of the initial curve is still high. In this case, the number ratio can be increased and smoothing can be performed again to further reduce the noise.

[0115] In some embodiments, when the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio includes: In the case where the number of peaks on the initial curve is smaller than the minimum value of the standard threshold range, the number ratio is reduced.

[0116] In this embodiment, when the number of peaks on the initial curve is less than the minimum value of the standard threshold range, for example, if a large difference is observed from the original data based on the initial curve, the number ratio can be appropriately reduced to make the production curve closer to the original data.

[0117] Thus, if the number of peaks on the initial curve is less than the minimum value of the standard threshold range, it can be considered that the initial curve is over-smoothed. In this case, the number ratio can be reduced and smoothing can be performed again to obtain a more accurate production curve.

[0118] Based on the production data anomaly detection method provided in the above embodiment, the present application also provides an embodiment of a production data anomaly detection device.

[0119] Figure 6 The schematic diagram of the structure of the production data anomaly detection device provided in the embodiment of the present application only shows the part related to the embodiment of the present application for the convenience of explanation.

[0120] Reference Figure 6 , the production data anomaly detection device 600 may include: An acquisition module 601 is used to acquire a production curve corresponding to the production data of the product to be tested, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value; A statistics module 602 is used to count the number of peaks on the production curve; The detection module 603 is used to perform anomaly detection based on the number of peaks.

[0121] In this embodiment, a production curve corresponding to the production data of the product to be tested can be obtained. The production curve can have a set of production specification parameter values ​​as the horizontal axis and a set of frequencies corresponding to each production specification parameter value as the vertical axis. The frequency of occurrence of the same production specification parameter value can be reflected, thereby counting the number of peaks on the production curve. Generally speaking, in normal production processes, the number of peaks is relatively stable, and slight changes in the distribution of production data can be reflected in the production curve, and the number of peaks will also change accordingly. Therefore, whether the production is abnormal can be judged based on whether the number of peaks changes, which can more comprehensively monitor the production situation and improve the accuracy of production data anomaly detection.

[0122] In some embodiments, the detection module 603 is further configured to: When the number of peaks is greater than a first threshold, it is determined that the production data of the product to be detected is abnormal, and the first threshold is a critical value of the number of peaks that characterizes the abnormal production data.

[0123] In this way, the number of peaks of the production curve corresponding to the production data of a certain production line or a certain device is counted, and based on the set first threshold, it is determined whether the production data of the production line or the device is abnormal, thereby realizing single-point dimensional production data anomaly detection.

[0124] In some embodiments, the production curve includes production curves corresponding to multiple production cycles, and the detection module 603 is further used to: Compare the number of peaks on the production curve corresponding to multiple production cycles; When the number of peaks on the production curves corresponding to the plurality of production cycles changes, it is determined that the production data of the product to be detected is abnormal.

[0125] In this way, for the production curves corresponding to multiple production cycles, the number of peaks in each production cycle can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production cycles shows a significant change, it can be considered that the distribution of production data in multiple production cycles is abnormal, thereby realizing production data anomaly detection in the time dimension.

[0126] In some embodiments, the production curve includes production curves corresponding to multiple production lines, and the detection module 603 is further used to: Compare the number of peaks on the production curves corresponding to multiple production lines; When the numbers of peaks on the production curves corresponding to the multiple production lines are not equal, it is determined that the production data of the product to be detected is abnormal.

[0127] In this way, for the production curves corresponding to multiple production lines, the number of peaks of each production line can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple production lines is unequal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0128] In some embodiments, the production curve includes production curves corresponding to multiple devices on the same production line, and the detection module 603 is further used to: Compare the number of peaks on the production curves corresponding to multiple devices; When the numbers of peaks on the production curves corresponding to the plurality of devices are not equal, it is determined that the production data of the product to be detected is abnormal.

[0129] In this way, for the production curves corresponding to multiple devices on the same production line, the number of peaks of each device can be counted. If it is detected that the number of peaks on the production curves corresponding to multiple devices is not equal, it can be considered that the consistency of the production data of the product to be tested is poor, thereby realizing production data anomaly detection in the spatial dimension.

[0130] In some embodiments, the production anomaly detection device 600 may further include: The determination module is used to determine the peaks on the production curve before counting the number of peaks on the production curve, wherein the ordinate of the peak is greater than the ordinates of other points in the neighborhood; wherein the distance between the abscissas of two adjacent peaks is greater than or equal to a distance threshold.

[0131] In this way, there will be only one peak in the same neighborhood, and there is a distance limit between two adjacent peaks, which can ensure the independence of each peak.

[0132] In some embodiments, the determination module is further configured to: Determine the initial point on the production curve, the ordinate of the initial point is greater than the ordinates of other points in the neighborhood; When the ordinate of the preliminary selected point is greater than or equal to the height threshold, the preliminary selected point is determined as the peak on the production curve.

[0133] Thus, in order to further improve the accuracy and relevance of peak identification, a height threshold is introduced. Only when the ordinate exceeds the height threshold, the corresponding preliminary point is officially identified as a peak.

[0134] In some embodiments, the acquisition module 601 further includes: An acquisition unit, used for acquiring production data of the product to be tested; A smoothing unit is used to smooth the production data; The generating unit is used to generate a production curve based on the smoothed production data.

[0135] In this way, the production data can be smoothed to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0136] In some embodiments, the smoothing unit is further configured to: For each data point in the production data, determine a neighborhood size of the data point based on a ratio of the number of data points; Assigning weights to all data points in the neighborhood, wherein the weight of the first data point is greater than the weight of the second data point, and the distance between the first data point and the center point in the neighborhood is less than the distance between the second data point and the center point; A local weighted linear regression operation is performed according to the weights of all data points in the neighborhood to obtain the smoothed coordinates of the data points.

[0137] In this way, the neighborhood size can be determined based on the parameter value, and the production data can be smoothed based on the local weighted linear regression operation to reduce noise, improve data quality, make the production curve more accurate, and thus improve the accuracy of production data anomaly detection.

[0138] In some embodiments, the smoothing unit is further configured to: Construct a weighted matrix based on the weights of all data points in the neighborhood; According to the weighted matrix and the linear regression equation, a local weighted linear regression equation is determined; Solve the model parameters in the locally weighted linear regression equation; Determine the smoothed coordinates of the data points based on the model parameters and the coordinates of the data points.

[0139] In this way, a local weighted linear regression equation can be constructed. After solving the model parameters, the smooth coordinates are calculated based on the model parameters and the coordinates of the data points to achieve the purpose of smoothing the data and improving the data quality.

[0140] In some embodiments, the smoothing unit is further configured to: Generate an initial curve based on the smoothed production data; Count the number of peaks on the initial curve; In the case where the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio; Smoothing of production data based on adjusted quantity ratios; When the number of peaks on the initial curve does not exceed a standard threshold range, the initial curve is determined as a production curve; The maximum and minimum values ​​of the standard threshold range are critical values ​​of the number of peaks that characterize the noise of the production data and meet the smoothing requirement.

[0141] In this way, the initial curve generated by the smoothed production data can be checked to determine whether the initial curve meets the requirements. If the number of peaks on the initial curve exceeds the standard threshold range, it can be considered that the initial curve does not meet the requirements. At this time, the number ratio can be adjusted to redetermine the neighborhood size and then smoothed, which further improves the accuracy of the production curve and thereby improves the accuracy of production data anomaly detection. If the number of peaks on the initial curve does not exceed the standard threshold range, it can be considered that the initial curve is a production curve that meets the requirements. At this time, accurate production data anomaly detection results can be obtained based on the production curve.

[0142] In some embodiments, the smoothing unit is further configured to: In the case where the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, the number ratio is increased.

[0143] Thus, if the number of peaks on the initial curve is greater than the number ratio of the standard threshold range, it can be considered that the noise of the initial curve is still high. In this case, the parameter value can be increased and smoothing can be performed again to further reduce the noise.

[0144] In some embodiments, the smoothing unit is further configured to: In the case where the number of peaks on the initial curve is smaller than the minimum value of the standard threshold range, the number ratio is reduced.

[0145] Thus, if the number of peaks on the initial curve is less than the minimum value of the standard threshold range, it can be considered that the initial curve is over-smoothed. In this case, the number ratio can be reduced and smoothing can be performed again to obtain a more accurate production curve.

[0146] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application, and are devices corresponding to the above-mentioned production data anomaly detection method. All implementation methods in the above-mentioned method embodiment are applicable to the embodiments of the device. Its specific functions and technical effects can be found in the method embodiment part, which will not be repeated here.

[0147] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0148] An embodiment of the present application also provides a battery production line, which uses the above-mentioned production data anomaly detection method to perform anomaly detection on production data.

[0149] In this way, whether the battery production is abnormal can be judged based on whether the number of peaks changes, so that the battery production situation can be monitored more comprehensively and the accuracy of abnormality detection in battery production data can be improved.

[0150] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0151] The electronic device 700 may include a processor 701 and a memory 702 storing programs or instructions. When the processor 701 executes the program, the steps in any of the above method embodiments are implemented.

[0152] Exemplarily, the program may be divided into one or more modules / units, one or more modules / units are stored in the memory 702 and executed by the processor 701 to complete the present application. One or more modules / units may be a series of program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the program in the device.

[0153] Specifically, the processor 701 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0154] The memory 702 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 702 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 702 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 702 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 702 is a non-volatile solid-state memory.

[0155] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0156] The processor 701 implements any one of the methods in the above embodiments by reading and executing the program or instruction stored in the memory 702 .

[0157] In one example, the electronic device may further include a communication interface 703 and a bus 704. The processor 701, the memory 702, and the communication interface 703 are connected via the bus 704 and communicate with each other.

[0158] The communication interface 703 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0159] Bus 704 includes hardware, software or both, and the components of online data traffic billing equipment are coupled to each other. For example, but not limitation, the bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. Where appropriate, bus 704 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0160] In addition, in combination with the method in the above embodiment, the embodiment of the present application may provide a machine-readable storage medium for implementation. The machine-readable storage medium stores a program or instruction; when the program or instruction is executed by a processor, any one of the methods in the above embodiment is implemented. The machine-readable storage medium can be read by a machine such as a computer.

[0161] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0162] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0163] An embodiment of the present application provides a computer program product, which is stored in a machine-readable storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0164] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0165] The functional modules shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer grid such as the Internet, an intranet, etc.

[0166] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0167] Aspects of the present disclosure are described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by a computer program or instruction. These programs or instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the functions / actions specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0168] Although the present application has been described with reference to preferred embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present application. In particular, the various technical features mentioned in the various embodiments may be combined in any manner as long as there is no structural conflict. The present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for detecting anomalies in production data, characterized in that: include: Obtaining a production curve corresponding to production data of the product to be tested, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value; Counting the number of peaks on the production curve; Anomaly detection is performed based on the number of peaks.

2. The method according to claim 1, characterized in that The performing anomaly detection based on the number of peaks includes: When the number of the peaks is greater than a first threshold, it is determined that the production data of the product to be detected is abnormal, and the first threshold is a critical value of the number of peaks that characterizes the abnormal production data.

3. The method according to claim 1 or 2, characterized in that: The production curve includes production curves corresponding to a plurality of production cycles, and the abnormality detection based on the number of the peaks includes: Comparing the numbers of peaks on the production curves corresponding to the multiple production cycles; When the number of peaks on the production curves corresponding to the multiple production cycles changes, it is determined that the production data of the product to be detected is abnormal.

4. The method according to claim 1 or 2, characterized in that: The production curve includes production curves corresponding to a plurality of production lines, and the abnormality detection based on the number of the peaks includes: Comparing the numbers of peaks on the production curves corresponding to the multiple production lines; When the numbers of peaks on the production curves corresponding to the multiple production lines are not equal, it is determined that the production data of the product to be detected is abnormal.

5. The method according to claim 4, characterized in that The production curve includes production curves corresponding to a plurality of devices on the same production line, and the abnormality detection based on the number of peaks includes: comparing the numbers of peaks on the production curves corresponding to the plurality of devices; When the numbers of peaks on the production curves corresponding to the plurality of devices are not equal, it is determined that the production data of the product to be detected is abnormal.

6. The method according to claim 1, characterized in that Before counting the number of peaks on the production curve, the method further includes: A peak on the production curve is determined, wherein the ordinate of the peak is greater than the ordinates of other points in the neighborhood; wherein the distance between the abscissas of two adjacent peaks is greater than or equal to a distance threshold.

7. The method according to claim 6, characterized in that Determining the peak on the production curve includes: Determine a preliminary point on the production curve, wherein the ordinate of the preliminary point is greater than the ordinates of other points in the neighborhood; When the ordinate of the preliminary selected point is greater than or equal to the height threshold, the preliminary selected point is determined as a peak on the production curve.

8. The method according to claim 1, characterized in that The step of obtaining a production curve corresponding to the production data of the product to be tested includes: Obtain production data of the product to be tested; performing smoothing processing on the production data; Generate a production curve based on the smoothed production data.

9. The method according to claim 8, characterized in that The smoothing of the production data comprises: For each data point in the production data, determining a neighborhood size of the data point based on a quantity ratio of the data points; Assigning weights to all data points in the neighborhood, wherein the weight of a first data point is greater than the weight of a second data point, and the distance between the first data point and a center point in the neighborhood is less than the distance between the second data point and the center point; A local weighted linear regression operation is performed according to the weights of all data points in the neighborhood to obtain smoothed coordinates of the data points.

10. The method according to claim 9, characterized in that The method of performing a local weighted linear regression operation according to the weights of all data points in the neighborhood to obtain the smoothed coordinates of the data points includes: Constructing a weighted matrix according to the weights of all data points in the neighborhood; Determine a local weighted linear regression equation according to the weighted matrix and the linear regression equation; Solving the model parameters in the local weighted linear regression equation; The smoothed coordinates of the data points are determined according to the model parameters and the coordinates of the data points.

11. The method according to claim 9 or 10, characterized in that: The method further comprises: Generate an initial curve based on the smoothed production data; Counting the number of peaks on the initial curve; When the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio, and performing smoothing processing on the production data based on the adjusted number ratio; When the number of peaks on the initial curve does not exceed the standard threshold range, determining the initial curve as the production curve; The maximum value and the minimum value of the standard threshold range are critical values ​​of the number of peaks that characterize the noise of the production data meeting the smoothing requirement.

12. The method according to claim 11, characterized in that When the number of peaks on the initial curve exceeds a standard threshold range, adjusting the number ratio comprises: In the case where the number of peaks on the initial curve is greater than the maximum value of the standard threshold range, increasing the number ratio; In the case where the number of peaks on the initial curve is smaller than the minimum value of the standard threshold range, the number ratio is reduced.

13. A production data anomaly detection device, characterized in that: include: An acquisition module, used to acquire a production curve corresponding to the production data of the product to be detected, wherein the horizontal axis of the production curve is used to indicate a set of production specification parameter values, and the vertical axis of the production curve is used to indicate a set of frequencies corresponding to each production specification parameter value; A statistical module, used for counting the number of peaks on the production curve; A detection module is used to perform anomaly detection based on the number of peaks.

14. A battery production line, characterized in that: The battery production line uses the production data anomaly detection method as described in any one of claims 1-12 to perform anomaly detection on production data.

15. An electronic device, characterized in that: The device comprises: a processor and a memory storing programs or instructions; When the processor executes the program or instruction, the method according to any one of claims 1 to 12 is implemented.

16. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

17. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method according to any one of claims 1 to 12.

Images