Welding quality evaluation method and device, electronic equipment and storage medium

Abstract

The invention discloses a welding quality evaluation method and device, electronic equipment and a storage medium. The welding quality evaluation method comprises the steps that parameter vectors of a welding seam are obtained, and the parameter vectors comprise parameter values of multiple welding characteristic parameters of the welding seam; obtaining a vector distance between the parameter vector and a historical sample vector set; each vector in the historical sample vector set comprises historical sample values of the various welding characteristic parameters; and based on the vector distance, whether the welding quality of the welding seam is abnormal or not is evaluated. According to the welding quality evaluation method, analysis of various parameters of the welding seam is achieved, comprehensive evaluation of the welding seam quality is achieved, the accuracy of the evaluation result is improved, the probability that potential welding quality problems are ignored is greatly reduced, and the requirement for welding quality monitoring can be better met.

Description

Technical Field

[0001] The present application relates to the field of welding technology, and in particular to a welding quality evaluation method, evaluation device, electronic equipment and storage medium. Background Art

[0002] With the rapid development of electric vehicles and renewable energy storage systems, the demand for lithium-ion batteries is growing. This type of battery is widely used in various fields due to its high energy density, long cycle life and good safety. However, the performance and safety of lithium-ion batteries are affected by many factors, among which the welding quality of welding operations in the battery manufacturing process is particularly important for battery performance. The welding quality is directly related to the overall performance of the battery, such as charging and discharging efficiency, cycle life, etc., and may even cause safety hazards such as short circuit or thermal runaway in extreme cases. Therefore, ensuring welding quality and conducting effective testing has become an important link that cannot be ignored in the manufacturing process of lithium-ion batteries.

[0003] In the related art, in the weld quality monitoring, only a certain basic parameter of the weld is analyzed, which limits the comprehensive evaluation of the weld quality and easily leads to some potential welding quality problems being ignored.

[0004] The above statements are only used to provide background information related to the present application and do not necessarily constitute prior art. Summary of the invention

[0005] In view of the fact that the above-mentioned related technologies only analyze a certain basic parameter of the weld in the weld quality monitoring, which limits the comprehensive evaluation of the weld quality and easily leads to the problem that some potential welding quality problems are ignored, the present application provides a welding quality evaluation method, evaluation device, electronic device and storage medium to realize the analysis of multiple parameters of the weld, realize a more comprehensive evaluation of the weld quality, greatly reduce the probability of potential welding quality problems being ignored, and better meet the needs of welding quality monitoring.

[0006] A first aspect of an embodiment of the present application provides a method for evaluating welding quality, comprising:

[0007] Acquire a parameter vector of a weld, wherein the parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld;

[0008] Acquire a vector distance between the parameter vector and a historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of the plurality of welding characteristic parameters;

[0009] Based on the vector distance, it is evaluated whether the welding quality of the weld is abnormal.

[0010] The welding quality evaluation method of the embodiment of the present application obtains the vector distance between the parameter vector and the historical sample vector set, and evaluates whether the welding quality of the weld is abnormal based on the vector distance. The parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld, thereby realizing analysis of multiple parameters of the weld, realizing a more comprehensive evaluation of the weld quality, improving the accuracy of the evaluation results, greatly reducing the probability of potential welding quality problems being ignored, and being able to better meet the needs of welding quality monitoring.

[0011] In some embodiments of the present application, the step of evaluating whether the welding quality of the weld is abnormal based on the vector distance includes:

[0012] When the vector distance does not belong to the preset distance interval, determining that the welding quality of the weld is abnormal;

[0013] When the vector distance belongs to the preset distance interval, it is determined that the welding quality of the weld is normal.

[0014] In this way, a more comprehensive assessment of weld quality is achieved, the accuracy of the assessment results is improved, the probability of potential welding quality problems being ignored is greatly reduced, and the needs of welding quality monitoring can be better met.

[0015] In some embodiments of the present application, the method further includes:

[0016] When the vector distance does not belong to the preset distance interval, a prompt message to stop welding is issued to remind relevant personnel to stop welding in time, reduce the generation of welds with abnormal welding quality, find the cause of abnormal welding quality in time and make adjustments to improve welding quality.

[0017] In some embodiments of the present application, the method further includes:

[0018] Obtaining a current exponentially weighted moving average value, where the current exponentially weighted moving average value is an exponentially weighted moving average value of the most recently obtained vector distance;

[0019] Based on the obtained preset number of exponentially weighted moving averages, obtaining a rate of change of the exponentially weighted moving average;

[0020] When the change rate does not belong to the preset change rate range, an early warning message is issued to indicate that there is a potential risk in welding quality.

[0021] In this way, potential trends of poor welding quality can be discovered in time, early warning of abnormal welding quality can be given, and the stability of welding quality can be improved.

[0022] In some embodiments of the present application, obtaining the current exponentially weighted moving average value includes:

[0023] Based on the preset smoothing factor, the most recently obtained vector distance and the last exponentially weighted moving average, the current exponentially weighted moving average is obtained. In this way, the current exponentially weighted moving average can be obtained more accurately to detect whether there is a potential poor welding quality condition.

[0024] In some embodiments of the present application, obtaining the current exponentially weighted moving average value based on a preset smoothing factor, the most recently obtained vector distance, and a previous exponentially weighted moving average value includes:

[0025] The ratio of the first value to the second value is calculated to obtain the current exponentially weighted moving average value; the first value is the sum of the third value and the fourth value, the second value is the difference between 1 and the preset smoothing factor, the third value is the product of the preset smoothing factor and the previous exponentially weighted moving average value, and the fourth value is the product of the second value and the most recently obtained vector distance. In this way, the current exponentially weighted moving average value can be calculated more accurately to detect whether there is a potential poor welding quality condition.

[0026] In some embodiments of the present application, obtaining the rate of change of the exponentially weighted moving average value based on the obtained preset number of exponentially weighted moving average values ​​includes:

[0027] Using a first sliding window to process the preset number of exponentially weighted moving averages, and sorting the data in the first sliding window in ascending order to obtain a sequence;

[0028] Obtaining a preset percentile of the sequence;

[0029] The preset percentiles of the sequence are linearly fitted in a second sliding window to obtain the slope of the fitting function, where the slope is the rate of change. In this way, the rate of change of the exponentially weighted moving average can be obtained more accurately to detect whether there is a potential risk in welding quality.

[0030] In some embodiments of the present application, obtaining the vector distance between the parameter vector and the historical sample vector set includes:

[0031] Obtain the mean value of the historical sample values ​​of each welding characteristic parameter;

[0032] According to the mean of the historical sample values ​​of each welding characteristic parameter, a covariance matrix and a deviation vector of the parameter vector are obtained;

[0033] The vector distance between the parameter vector and the historical sample vector set is calculated according to the covariance matrix and the deviation vector.

[0034] In this way, the analysis of multiple parameters of the weld is achieved, which helps to make a more comprehensive assessment of the weld quality and helps to improve the accuracy of the assessment results.

[0035] In some embodiments of the present application, obtaining the parameter vector of the weld includes:

[0036] Obtaining the length, width and height of the weld;

[0037] The parameter vector of the weld is constructed using the length, the width and the height as elements.

[0038] By acquiring multiple parameters of the weld, it is helpful to analyze the multiple parameters of the weld, to make a more comprehensive evaluation of the weld quality, and to improve the accuracy of the evaluation results.

[0039] In some embodiments of the present application, the vector distance includes Mahalanobis distance or Euclidean distance.

[0040] The Mahalanobis distance can well reflect the correlation between various welding characteristic parameters, thereby providing a more accurate similarity calculation, better reflecting the similarity of data points, and helping to improve the accuracy of the evaluation results.

[0041] According to a second aspect of an embodiment of the present application, there is provided a welding quality assessment device, comprising:

[0042] A parameter vector acquisition module, used to acquire a parameter vector of a weld, wherein the parameter vector includes parameter values ​​of a plurality of welding characteristic parameters of the weld;

[0043] A vector distance acquisition module, used to acquire the vector distance between the parameter vector and a historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of the multiple welding characteristic parameters;

[0044] An evaluation module is used to evaluate whether the welding quality of the weld is abnormal based on the vector distance.

[0045] The welding quality evaluation device of the embodiment of the present application can obtain the vector distance between the parameter vector and the historical sample vector set, and evaluate whether the welding quality of the weld is abnormal based on the vector distance. The parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld, thereby realizing analysis of multiple parameters of the weld, realizing a more comprehensive evaluation of the weld quality, improving the accuracy of the evaluation results, greatly reducing the probability of potential welding quality problems being ignored, and being able to better meet the needs of welding quality monitoring.

[0046] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the welding quality evaluation method described in any embodiment of the present application.

[0047] The electronic device provided in the third aspect of the embodiment of the present application can achieve the beneficial technical effect that can be achieved by the method provided in the first aspect of the embodiment of the present application.

[0048] According to a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program is executed by a processor to implement the welding quality evaluation method described in any embodiment of the present application.

[0049] The computer-readable storage medium provided in the fourth aspect of the embodiments of the present application can achieve the beneficial technical effects that can be achieved by the method provided in the first aspect of the embodiments of the present application.

[0050] The above description is only an overview of the technical solution of the embodiment of the present application. In order to more clearly understand the technical means of the embodiment 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 embodiment 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

[0051] By reading the detailed description of the following embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The accompanying drawings are only used to illustrate the embodiments of the present application and are not to be considered as limitations of the present application. In addition, the same reference numerals are used to represent the same components throughout the drawings.

[0052] Figure 1 The present invention is a flow chart of a method for evaluating welding quality according to one or more embodiments.

[0053] Figure 2 A flowchart of obtaining a vector distance between a parameter vector and a set of historical sample vectors according to one or more embodiments.

[0054] Figure 3 is a graph of multiple Mahalanobis distances according to one or more embodiments.

[0055] Figure 4 is a graph of multiple exponentially weighted moving averages according to one or more embodiments.

[0056] Figure 5 A graph of percentiles of multiple exponentially weighted moving averages according to one or more embodiments.

[0057] Figure 6 The figure is a structural block diagram of a device for evaluating welding quality according to one or more embodiments.

[0058] Figure 7 is a structural block diagram of an electronic device according to one or more embodiments.

[0059] Figure 8 FIG. 1 is a schematic diagram of a computer-readable storage medium according to one or more embodiments. DETAILED DESCRIPTION

[0060] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0062] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two (including two), unless otherwise clearly and specifically defined.

[0063] Reference to "embodiments" herein 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 herein may be combined with other embodiments.

[0064] With the rapid development of new energy technologies, batteries, especially power batteries, have been widely used in more and more technical fields. Power batteries are widely used in many fields including electric vehicles due to their high energy density, long cycle life and good safety. The performance and safety of batteries are affected by many factors, among which the welding quality of welding operations in the battery manufacturing process is particularly important to the battery performance.

[0065] The welding quality in the battery manufacturing process is directly related to the overall performance of the battery, such as charging and discharging efficiency, cycle life, etc., and may even cause safety hazards such as short circuit or thermal runaway in extreme cases. Therefore, ensuring welding quality and conducting effective testing has become an important link that cannot be ignored in the lithium-ion battery manufacturing process.

[0066] In the related art, in the weld quality monitoring, only a certain basic parameter of the weld is analyzed, which limits the comprehensive evaluation of the weld quality and easily leads to some potential welding quality problems being ignored.

[0067] In response to the problems existing in the related technology, an embodiment of the present application provides a welding quality assessment method, which obtains the vector distance between a parameter vector and a historical sample vector set, and evaluates whether the welding quality of the weld is abnormal based on the vector distance. The parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld, thereby realizing analysis of multiple parameters of the weld, achieving a more comprehensive assessment of the weld quality, improving the accuracy of the assessment results, greatly reducing the probability of potential welding quality problems being ignored, and being able to better meet the needs of welding quality monitoring.

[0068] The welding quality evaluation method provided in the embodiment of the present application can be applied to various application scenarios including welding process steps. For example, it can be applied to the evaluation of welding quality in battery welding process steps in the battery manufacturing process, such as the evaluation of Busbar welding quality.

[0069] A welding quality evaluation method, evaluation device, electronic device, and storage medium proposed according to an embodiment of the present application are described below in conjunction with the accompanying drawings.

[0070] refer to Figure 1 As shown, an embodiment of the present application provides a method for evaluating welding quality, which may include steps S10 to S30:

[0071] S10. Obtain a parameter vector of the weld, where the parameter vector includes parameter values ​​of various welding characteristic parameters of the weld.

[0072] Specifically, the various welding characteristic parameters of the weld may include but are not limited to weld length, weld width and weld height. During the welding process, the weld can be monitored in real time by a high-precision sensor or a CCD camera to collect parameter values ​​of the various welding characteristic parameters of the weld. These parameter values ​​will serve as the basis for subsequent analysis to facilitate a more comprehensive evaluation of the welding quality. CCD is the abbreviation for charge coupled device. Welding includes but is not limited to Busbar welding.

[0073] In some implementations, obtaining a parameter vector of a weld may include:

[0074] S101. Obtain the length, width and height of the weld.

[0075] Specifically, a high-precision CCD camera is used to photograph the weld to capture a complete image of the weld. The outer edge contour of the weld is extracted through an image processing algorithm, and the length of the outermost edge of the weld is calculated. This step can more accurately obtain the actual length of the weld and achieve coverage of the welding area.

[0076] Use a CCD camera to obtain the weld image and select multiple measurement points. By analyzing the distance between the outermost and innermost edges of the weld, the width of each point is calculated and the average is taken to obtain the weld width. This measurement can better reflect the overall consistency of the weld.

[0077] The grayscale analysis of the image is used to establish the relationship between the grayscale and height of the weld area. By measuring the grayscale changes in the weld area, applying the set threshold, judging the height of the weld, and recording its average value. This process helps to identify possible quality problems in the welding process.

[0078] The above welding characteristic parameters are recorded in real time and stored synchronously with the welding parameters (such as laser power, welding speed, etc.) to provide comprehensive data support for subsequent analysis.

[0079] S102. Use length, width and height as elements to construct a parameter vector of the weld.

[0080] The welding characteristic parameters of the weld are used to form a parameter vector including weld length, weld width and weld height.

[0081] By acquiring multiple parameters of the weld, it is helpful to analyze the multiple parameters of the weld, to make a more comprehensive evaluation of the weld quality, and to improve the accuracy of the evaluation results.

[0082] Exemplarily, the vector distance includes Mahalanobis distance or Euclidean distance. Mahalanobis distance can well reflect the correlation between various welding characteristic parameters, thereby providing more accurate similarity calculation, better reflecting the similarity of data points, and helping to improve the accuracy of evaluation results.

[0083] S20, obtaining the vector distance between the parameter vector and the historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of multiple welding characteristic parameters.

[0084] In some embodiments, the vector distance is the Mahalanobis distance; Figure 2 As shown, obtaining the vector distance between the parameter vector and the historical sample vector set may include steps S201 to S203:

[0085] S201. Obtain the mean of the historical sample values ​​of each welding characteristic parameter.

[0086] For example, the mean value of the weld characteristic parameter is calculated by using the historical sample data collected in advance, where the historical sample data is the welding characteristic parameter value collected when the welding quality is normal. Where X represents the weld characteristic parameter, x represents the historical sample value corresponding to the weld characteristic parameter, n represents the amount of historical data, i represents the historical data index, and E represents the expectation. The mean value can reflect the typical characteristics of a normal weld.

[0087] S202. Obtain a covariance matrix and a deviation vector of a parameter vector according to the mean of the historical sample values ​​of each welding characteristic parameter.

[0088] For example, the covariance matrix (Take two-dimensional features as an example), and the covariance matrix can provide the correlation information between each welding feature parameter.

[0089] A mean vector is constructed using the mean of each weld characteristic parameter, and the deviation vector is obtained by subtracting the mean vector from the parameter vector.

[0090] For example, the parameter vector is (the currently detected weld length, the currently detected weld width, the currently detected weld height), the mean vector is (the historical weld length mean, the historical weld width mean, the historical weld height mean), and the deviation vector is (the difference between the currently detected weld length and the historical weld length mean, the difference between the currently detected weld width and the historical weld width mean, the difference between the currently detected weld height and the historical weld height mean).

[0091] S203. Calculate the vector distance between the parameter vector and the historical sample vector set according to the covariance matrix and the deviation vector.

[0092] Taking the case where the vector distance is the Mahalanobis distance as an example, firstly, the transposed matrix of the deviation vector is calculated, and then the transposed matrix, the inverse matrix of the covariance matrix and the deviation vector are multiplied in sequence to obtain a scalar value, and the square root of the scalar value is calculated to obtain the Mahalanobis distance.

[0093] In this way, the analysis of multiple parameters of the weld is achieved, which helps to make a more comprehensive assessment of the weld quality and helps to improve the accuracy of the assessment results.

[0094] In some implementations, obtaining the Mahalanobis distance between the parameter vector and the historical sample vector set may further include:

[0095] According to the mean value of weld characteristic parameters And the covariance matrix Calculate the Mahalanobis distance between the parameter vector and the historical sample vector set Where j represents the index of the online detection data. The Mahalanobis distance can be used to determine the degree of deviation between the current weld characteristic parameters and the weld characteristic parameters in the normal state, thereby identifying possible defects and marking the detected abnormal values ​​as abnormal points.

[0096] S30. Based on the vector distance, assess whether the welding quality of the weld is abnormal.

[0097] In some embodiments, evaluating whether the welding quality of the weld is abnormal based on the vector distance may include: when the vector distance does not belong to a preset distance interval, determining that the welding quality of the weld is abnormal; when the vector distance belongs to the preset distance interval, determining that the welding quality of the weld is normal.

[0098] In this way, a more comprehensive assessment of weld quality is achieved, the accuracy of the assessment results is improved, the probability of potential welding quality problems being ignored is greatly reduced, and the needs of welding quality monitoring can be better met.

[0099] In some embodiments, the method may further include:

[0100] S40. When the vector distance does not belong to the preset distance interval, a prompt message to stop welding is issued to remind relevant staff to stop welding in time, reduce the generation of welds with abnormal welding quality, find out the cause of abnormal welding quality in time and make adjustments to improve welding quality.

[0101] Exemplarily, when the vector distance belongs to a preset distance interval, the parameter vector of the weld is obtained and the process is repeated.

[0102] In some embodiments, the method may further include steps S50 to S70:

[0103] S50: Obtain a current exponentially weighted moving average value, where the current exponentially weighted moving average value is an exponentially weighted moving average value of the most recently obtained vector distance.

[0104] In some implementations, obtaining the current exponentially weighted moving average value may include:

[0105] Based on the preset smoothing factor, the most recently obtained vector distance and the last exponentially weighted moving average value, the current exponentially weighted moving average value is obtained. In this way, the current exponentially weighted moving average value can be obtained more accurately to detect whether there is a potential poor welding quality condition.

[0106] Exemplarily, obtaining the current exponentially weighted moving average value based on a preset smoothing factor, a most recently obtained vector distance, and a previous exponentially weighted moving average value may include:

[0107] The ratio of the first value to the second value is calculated to obtain the current exponentially weighted moving average value; the first value is the sum of the third value and the fourth value, the second value is the difference between 1 and the preset smoothing factor, the third value is the product of the preset smoothing factor and the previous exponentially weighted moving average value, and the fourth value is the product of the second value and the most recently obtained vector distance. In this way, the current exponentially weighted moving average value can be calculated more accurately to detect whether there is a potential poor welding quality condition.

[0108] In another embodiment, a suitable smoothing factor λ is first determined to capture the short-term and long-term variation trends of the Mahalanobis distance. After each welding is completed, the current Mahalanobis distance value is input into the EWMA algorithm in real time to calculate the EWMA at the current moment. where v t-1 Indicates the EWMA of the previous moment, D t It represents the Mahalanobis distance at the current moment. From the formula, we can know that the EWMA at the current moment can be calculated only by the EWMA at the previous moment and the Mahalanobis distance at the current moment. It does not need to save all the past data, and has a low demand for computing memory, which is suitable as an online detection algorithm.

[0109] S60 . Based on a preset number of exponentially weighted moving average values ​​obtained, obtain a rate of change of the exponentially weighted moving average value.

[0110] Exemplarily, obtaining the rate of change of the exponentially weighted moving average value based on the obtained preset number of exponentially weighted moving average values ​​may include:

[0111] S601, using a first sliding window to process a preset number of exponentially weighted moving averages, and sorting the data in the first sliding window in ascending order to obtain a sequence;

[0112] S602, obtaining a preset percentile of the sequence;

[0113] S603, perform linear fitting of the second sliding window on the preset percentiles of the sequence to obtain the slope of the fitting function, where the slope is the rate of change. In this way, the rate of change of the exponentially weighted moving average can be obtained more accurately to detect whether there is a potential risk in welding quality.

[0114] The size of the first sliding window and the size of the second sliding window can be selected according to the needs of the actual application. For example, the size of the sliding window can be set to 50 data points, and the preset percentile can be selected according to the needs of the actual application, such as the 99% percentile.

[0115] In another embodiment, the EWMA trend is continuously monitored, and the EWMA results are processed using sliding windows and percentiles to avoid the influence of extreme outliers on trend judgment. Sort the data in the sliding window from small to large and find Q p The number of the position is the result of the pth percentile. If Q p If the result is not an integer, it is necessary to use linear interpolation to estimate. Perform linear fitting of the sliding window on the percentile result to obtain the slope of the fitting function, and use the least squares method to calculate the slope of the fitting function. Where m represents the size of the sliding window. When k is greater than or less than a preset threshold, it is marked as a warning point. When the warning point is reached, there is a potential risk in welding quality.

[0116] S70: When the change rate does not fall within the preset change rate range, a warning message is issued to indicate that there is a potential risk to welding quality.

[0117] In this way, potential trends of poor welding quality can be discovered in time, early warning of abnormal welding quality can be given, and the stability of welding quality can be improved.

[0118] In some examples, when a warning point is reached, a warning message may be issued to indicate that there is a potential risk to welding quality.

[0119] The related technology has a weak ability to detect anomalies, and it is difficult to effectively identify sudden problems in the welding process, resulting in the inability to take corrective measures in time. In addition, the trend deviation of weld parameters is also difficult to be monitored in time in the detection method of the related technology. If a trend change occurs in a weld parameter, it is often impossible to perceive it in advance, so that the occurrence of poor welding conditions cannot be effectively prevented. This situation is particularly serious in large-scale production, which may lead to fluctuations in the overall product quality, and the occurrence of poor batch welding cannot be avoided, resulting in unnecessary cost waste. Therefore, improving the comprehensiveness and real-time nature of post-weld detection technology to achieve comprehensive monitoring of the welding process is also an urgent problem to be solved in the battery manufacturing industry. The Mahalanobis distance and the exponentially weighted moving average method are introduced into the welding quality evaluation method of the embodiment of the present application. The Mahalanobis distance is used to comprehensively analyze multiple parameters, and the exponentially weighted moving average rule is used to identify the trend deviation of the Mahalanobis distance, which can achieve a more accurate and comprehensive welding quality evaluation.

[0120] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0121] In a specific example, the welding quality assessment method may include steps 101 to 104:

[0122] Step 101: Data collection.

[0123] During the welding process, the weld is first monitored in real time by a high-precision sensor or CCD camera to collect relevant parameters of the weld. These data will serve as the basis for subsequent analysis to ensure a comprehensive assessment of the welding quality.

[0124] Busbar laser welding technology is one of the key steps in the battery manufacturing process. It is used to connect the connecting piece with the pole, thereby realizing the series and parallel connection between the battery cells and providing the necessary current channel. During the Busbar welding process, the key parameters of the weld are monitored in real time through a CCD camera. The specific steps are as follows:

[0125] Weld length: Use a high-precision CCD camera to photograph the weld and capture the complete image of the weld. Use image processing algorithms to extract the outer edge contour of the weld and calculate the length of the outermost edge of the weld. This step ensures that the actual length of the weld can be accurately reflected and the coverage of the welding area can be guaranteed.

[0126] Weld width: Also use the CCD camera to obtain the weld image and select multiple measurement points. By analyzing the distance between the outermost and innermost edges of the weld, the width of each point is calculated and the average is taken to obtain the weld width. This measurement can better reflect the overall consistency of the weld.

[0127] Weld Height: Use grayscale analysis of the image to establish the relationship between the grayscale and height of the weld area. By measuring the grayscale changes in the weld area, applying the set threshold, determine the height of the weld, and record its average value. This process helps identify possible quality problems in the welding process.

[0128] The above data are recorded in real time and stored synchronously with welding parameters (such as laser power, welding speed, etc.), providing comprehensive data support for subsequent analysis.

[0129] Step 102: Mahalanobis distance analysis.

[0130] The collected weld parameters are analyzed using the Mahalanobis distance method. This method can combine multiple parameters to calculate the comprehensive index of weld quality. By comparing with normal welding samples, the deviation of weld parameters can be identified, thus effectively judging the quality of welding.

[0131] Data preparation:

[0132] The characteristic parameters of the current weld are obtained from the data acquisition stage to form a vector containing length, width and height.

[0133] Mean and covariance calculation:

[0134] Collect historical normal sample data and calculate the mean value of weld characteristic parameters and the covariance matrix (Take two-dimensional features as an example), where X represents the weld feature parameters, x represents the historical data corresponding to the weld feature, n represents the amount of historical data, i represents the historical data index, and E represents the expectation. The mean reflects the typical characteristics of a normal weld, and the covariance matrix provides the correlation information between the features.

[0135] Mahalanobis distance calculation:

[0136] The Mahalanobis distance of the current sample is calculated by using the characteristic parameters of the current weld and the mean and covariance matrix of the historical normal samples. Where j represents the index of the online detection data. This distance can be used to determine the degree of deviation between the current weld characteristics and the normal state, thereby identifying possible defects and marking them as abnormal points when there are abnormal values.

[0137] Step 103: EWMA trend detection.

[0138] Based on the Mahalanobis distance analysis, the exponential weighted moving average method (EWMA) is used to monitor the trend of weld quality. By identifying the trend deviation of the Mahalanobis distance results, potential poor welding trends can be discovered in time. This method can predict the occurrence of poor welding in advance and ensure the stability of welding quality.

[0139] Based on the Mahalanobis distance analysis, EWMA trend detection is performed. The specific steps are as follows:

[0140] Initialization parameters: Determine the appropriate smoothing factor λ to capture the short-term and long-term variation trends of the Mahalanobis distance.

[0141] Real-time update: After each welding is completed, the current Mahalanobis distance value is input into the EWMA algorithm in real time to calculate the current moment EWMA. where υ t-1 Indicates the EWMA of the previous moment, D t It represents the Mahalanobis distance at the current moment. From the above formula, we can know that the EWMA at the current moment can be calculated only by the EWMA at the previous moment and the Mahalanobis distance at the current moment. It does not need to save all the past data, and has a low demand for computing memory, which is suitable as an online detection algorithm.

[0142] Trend analysis: Continuously monitor the EWMA trend, use sliding windows and percentiles to process the EWMA results to avoid the influence of extreme outliers on trend judgment, and the pth percentile position Sort the data in the sliding window from small to large and find Q pThe number of the position is the result of the pth percentile. If Q p If the result is not an integer, it is necessary to use linear interpolation to estimate. Perform linear fitting of the sliding window on the percentile result to obtain the slope of the fitting function, and use the least squares method to calculate the slope of the fitting function. Where m represents the size of the sliding window. When k is greater than or less than a preset threshold, it is marked as a warning point.

[0143] Step 104: Output and feedback of analysis results.

[0144] The analysis results are output through a visual interface and provided to relevant operators. If abnormal welding quality is detected, the system will automatically generate a feedback report or signal to guide the operator or equipment to make corresponding adjustments or inspections to improve the quality and stability of welding.

[0145] Generate an alarm signal based on the detection results, which can be divided into the following two situations:

[0146] (1) Abnormal Mahalanobis distance: If the Mahalanobis distance exceeds the set defect signal threshold, the system will issue an alarm signal to stop the production line for inspection and troubleshooting, prompting the operator to immediately stop the production line to inspect and rectify the defective welds to ensure that the welding quality meets the standards.

[0147] (2) Abnormal EWMA trend: If EWMA detects an abnormal trend, but the Mahalanobis distance does not exceed the defect signal threshold, the system will issue a warning signal. This signal reminds the operator that there is a potential risk in the current welding and that the operator needs to pay attention to the changes in welding parameters, but the machine does not need to be shut down immediately and can be judged and adjusted according to the actual situation.

[0148] Further, the example effect of applying this example method in actual Busbar welding is as follows: Figure 3 and Figure 4 shown. Figure 3 It is a curve diagram of Mahalanobis distance corresponding to multiple detection data. Each time a welding characteristic parameter detection is performed, a parameter vector is constructed using the length, width and height of the weld obtained by the detection. Each parameter vector corresponds to a Mahalanobis distance. Figure 3 The numbers on the horizontal axis represent the serial numbers of the parameter value data obtained in each test, and each test corresponds to a serial number. The vertical axis represents the value of the Mahalanobis distance. Figure 4 It is a graph of multiple exponentially weighted moving averages, where each Mahalanobis distance corresponds to an exponentially weighted moving average. Figure 4The numbers on the horizontal axis represent the parameter value data obtained in each test. Each test corresponds to a number, and the vertical axis represents the exponentially weighted moving average. In this example, the Busbar welding was broken at step 522 because the laser was blocked by the gas pipe. However, the upper and lower limits of the width, height and length of the weld failed to intercept it, resulting in missed judgment. Figure 3 It can be seen from the Mahalanobis distance detection results that the Mahalanobis distance corresponding to the data detected at this moment rises sharply, which obviously exceeds the normal level, and can be accurately intercepted to avoid abnormal leakage. At 800 to 3500 steps, due to the poor welding of the collimator mirror water cooling pipe, the collimator mirror leaks, which in turn causes the collimator mirror to become dirty, and the welding area receives insufficient energy, resulting in Busbar cold welding, which requires shutdown and disassembly for investigation, resulting in waste of production capacity and cost. Figure 5 A graph of percentiles for multiple exponentially weighted moving averages. Figure 5 The numbers on the horizontal axis represent the percentile numbers, each percentile corresponds to a number, and the vertical axis represents the exponentially weighted moving average. Figure 5 Several warning points are marked in Figure 5 From the distribution of the exponentially weighted moving average warning points, it can be seen that in the early stage of the risk, a warning signal is issued at about 1000 steps. By checking the abnormality according to the warning signal on site, the subsequent Busbar welding cold welds can be avoided, thereby effectively reducing the cost waste caused by unexpected downtime and disassembly.

[0149] In the method of this example, the Mahalanobis distance method is used to comprehensively analyze multiple parameters of the weld, and the overall quality of the weld is evaluated by calculating the Mahalanobis distance between the parameters. This method takes into account the covariance between the parameters and can effectively identify abnormal conditions of weld characteristics. Applying the EWMA method to monitor the trend of the Mahalanobis distance and obtaining the preset percentile of the EWMA value can assign higher weights to the most recent data, and identify abnormal deviations by dynamically analyzing the changing trend of the Mahalanobis distance.

[0150] In the method of this example, a sliding window (such as 50 data points) is set to calculate the percentile (such as 99% percentile) of the EWMA result within the window. The automated process design and data processing mechanism for weld quality inspection ensures real-time and accuracy.

[0151] In this example, the Mahalanobis distance method is used to comprehensively analyze the length, width and height of the weld during the welding process, effectively combining the relationship between multiple variables to obtain a comprehensive indicator that reflects the overall welding quality. Compared with the single parameter upper and lower limit method, the evaluation dimensions are more diverse and more accurate, avoiding a large number of misjudgments and omissions of welding quality problems.

[0152] In the method of this example, the results after comprehensive analysis are smoothed by an exponentially weighted moving average (EWMA) model to reduce the impact of random fluctuations in data on trend analysis, thereby more accurately monitoring the long-term trend of the data. When the trend deviates from the preset range, an alarm is triggered in time to reduce the risk of poor welding.

[0153] In the method of this example, the sliding window technology is applied to analyze the data processed by EWMA. The sliding window design allows the system to update and manage the latest data points in real time, ensuring that welding quality monitoring is always based on the most recent actual data. The sliding window mechanism automatically excludes historical data, effectively avoiding the unlimited growth of computing memory, and also reducing the impact of historical data on current analysis, thereby effectively reducing noise and interference, making trend analysis more accurate. The percentile of each window is calculated to detect abnormal fluctuations or trend changes in the data, which can effectively reduce the impact of occasional outliers on the overall trend analysis and enhance the stability of the algorithm. It is also possible to set multi-level abnormal thresholds so that the system can automatically identify continuous data decline or increase trends, and issue an alarm when an abnormality is detected, so as to intervene and correct in time, and finally realize the automation of the detection process, improve detection efficiency, and reduce manual intervention and misjudgment.

[0154] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0155] refer to Figure 6 As shown, another embodiment of the present application provides a welding quality assessment device, comprising:

[0156] A parameter vector acquisition module is used to acquire a parameter vector of a weld, wherein the parameter vector includes parameter values ​​of various welding characteristic parameters of the weld;

[0157] A vector distance acquisition module is used to acquire the vector distance between the parameter vector and the historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of multiple welding characteristic parameters;

[0158] The evaluation module is used to evaluate whether the welding quality of the weld is abnormal based on the vector distance.

[0159] Exemplarily, the vector distance includes Mahalanobis distance or Euclidean distance.

[0160] Exemplarily, the evaluation module is further specifically used to: when the vector distance does not belong to the preset distance interval, determine that the welding quality of the weld is abnormal; when the vector distance belongs to the preset distance interval, determine that the welding quality of the weld is normal.

[0161] Exemplarily, the parameter vector acquisition module is further specifically used to: acquire the length, width and height of the weld; and construct a parameter vector of the weld using the length, width and height as elements.

[0162] Exemplarily, the vector distance acquisition module is further specifically used to: obtain the mean of the historical sample values ​​of each welding characteristic parameter; obtain the covariance matrix and the deviation vector of the parameter vector based on the mean of the historical sample values ​​of each welding characteristic parameter; and calculate the vector distance between the parameter vector and the historical sample vector set based on the covariance matrix and the deviation vector.

[0163] Exemplarily, the device may further include:

[0164] The prompt module is used to issue a prompt message to stop welding when the vector distance does not belong to the preset distance interval.

[0165] Exemplarily, the device may further include:

[0166] The EWMA acquisition module is used to obtain the current exponentially weighted moving average value, which is the exponentially weighted moving average value of the most recently obtained vector distance;

[0167] A change rate acquisition module, used for acquiring a change rate of the exponentially weighted moving average value based on a preset number of exponentially weighted moving average values ​​obtained;

[0168] The early warning module is used to issue an early warning message indicating that there is a potential risk to welding quality when the change rate does not fall within a preset change rate range.

[0169] Exemplarily, the EWMA acquisition module is further specifically configured to acquire a current exponentially weighted moving average value based on a preset smoothing factor, a most recently obtained vector distance, and a previous exponentially weighted moving average value.

[0170] Exemplarily, the EWMA acquisition module is further specifically used to calculate the ratio of the first value to the second value to obtain the current exponentially weighted moving average; the first value is the sum of the third value and the fourth value, the second value is the difference between 1 and a preset smoothing factor, the third value is the product of the preset smoothing factor and the previous exponentially weighted moving average, and the fourth value is the product of the second value and the most recently obtained vector distance.

[0171] Exemplarily, the change rate acquisition module is further specifically used for:

[0172] Using a first sliding window to process a preset number of exponentially weighted moving averages, and sorting the data in the first sliding window in ascending order to obtain a sequence;

[0173] Get the preset percentile of a series;

[0174] Perform a linear fit of the second sliding window on the preset percentiles of the sequence to obtain the slope of the fitting function, which is the rate of change.

[0175] The welding quality evaluation device of the embodiment of the present application can obtain the vector distance between the parameter vector and the historical sample vector set, and evaluate whether the welding quality of the weld is abnormal based on the vector distance. The parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld, thereby realizing analysis of multiple parameters of the weld, realizing a more comprehensive evaluation of the weld quality, improving the accuracy of the evaluation results, greatly reducing the probability of potential welding quality problems being ignored, and being able to better meet the needs of welding quality monitoring.

[0176] Another embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method of any of the above embodiments.

[0177] refer to Figure 7 As shown, the electronic device 10 may include: a processor 100, a memory 101, a bus 102 and a communication interface 103, and the processor 100, the communication interface 103 and the memory 101 are connected via the bus 102; the memory 101 stores a computer program that can be run on the processor 100, and when the processor 100 runs the computer program, it executes the method provided in any of the aforementioned embodiments of the present application.

[0178] The memory 101 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the device network element and at least one other network element is realized through at least one communication interface 103 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0179] The bus 102 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 101 is used to store a program, and the processor 100 executes the program after receiving an execution instruction. The method disclosed in any implementation of the aforementioned embodiment of the present application may be applied to the processor 100, or implemented by the processor 100.

[0180] The processor 100 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 100 or an instruction in the form of software. The above processor 100 may be a general-purpose processor, which may include a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 101, and the processor 100 reads the information in the memory 101 and completes the steps of the above method in combination with its hardware.

[0181] The electronic device provided in the embodiment of the present application and the method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the method adopted, operated or implemented by them.

[0182] Another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the method of any of the above embodiments. Figure 8 As shown, the computer-readable storage medium shown is a CD 20 on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, the method provided by any of the aforementioned embodiments will be executed.

[0183] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0184] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0185] It should be noted that:

[0186] The term "module" is not intended to be limited to a specific physical form. Depending on the specific application, a module can be implemented as hardware, firmware, software, and / or a combination thereof. In addition, different modules can share common components or even be implemented by the same components. There may or may not be clear boundaries between different modules.

[0187] The algorithm and display provided herein are not inherently related to any specific computer, virtual device or other equipment. Various general devices can also be used together with examples based on this. According to the above description, it is obvious to construct the structure required for this type of device. In addition, the application is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the application described herein, and the description made to specific languages ​​above is for the purpose of disclosing the best mode of implementation of the application.

[0188] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0189] The above embodiments only express the implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for evaluating welding quality, characterized in that: include: Acquire a parameter vector of a weld, wherein the parameter vector includes parameter values ​​of multiple welding characteristic parameters of the weld; Acquire a vector distance between the parameter vector and a historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of the plurality of welding characteristic parameters; Based on the vector distance, it is evaluated whether the welding quality of the weld is abnormal.

2. The method according to claim 1, characterized in that The step of evaluating whether the welding quality of the weld is abnormal based on the vector distance includes: When the vector distance does not belong to the preset distance interval, determining that the welding quality of the weld is abnormal; When the vector distance belongs to the preset distance interval, it is determined that the welding quality of the weld is normal.

3. The method according to claim 2, characterized in that The method further comprises: When the vector distance does not belong to the preset distance interval, a prompt message to stop welding is issued.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Obtaining a current exponentially weighted moving average value, where the current exponentially weighted moving average value is an exponentially weighted moving average value of the most recently obtained vector distance; Based on the obtained preset number of exponentially weighted moving averages, obtaining a rate of change of the exponentially weighted moving average; When the change rate does not belong to the preset change rate range, an early warning message is issued to indicate that there is a potential risk in welding quality.

5. The method according to claim 4, characterized in that The method of obtaining the current exponentially weighted moving average value includes: The current exponentially weighted moving average value is obtained based on a preset smoothing factor, the most recently obtained vector distance, and a previous exponentially weighted moving average value.

6. The method according to claim 5, characterized in that The obtaining the current exponentially weighted moving average value based on the preset smoothing factor, the most recently obtained vector distance, and the last exponentially weighted moving average value comprises: Calculate the ratio of the first value to the second value to obtain the current exponentially weighted moving average; the first value is the sum of the third value and the fourth value, the second value is the difference between 1 and the preset smoothing factor, the third value is the product of the preset smoothing factor and the previous exponentially weighted moving average, and the fourth value is the product of the second value and the most recently obtained vector distance.

7. The method according to claim 4, characterized in that The step of obtaining the rate of change of the exponentially weighted moving average based on the obtained preset number of exponentially weighted moving averages includes: Using a first sliding window to process the preset number of exponentially weighted moving averages, and sorting the data in the first sliding window in ascending order to obtain a sequence; Obtaining a preset percentile of the sequence; A linear fit of a second sliding window is performed on a preset percentile of the sequence to obtain a slope of a fitting function, where the slope is the rate of change.

8. The method according to any one of claims 1 to 3, characterized in that The obtaining of the vector distance between the parameter vector and the historical sample vector set includes: Obtain the mean value of the historical sample values ​​of each welding characteristic parameter; According to the mean of the historical sample values ​​of each welding characteristic parameter, a covariance matrix and a deviation vector of the parameter vector are obtained; The vector distance between the parameter vector and the historical sample vector set is calculated according to the covariance matrix and the deviation vector.

9. The method according to any one of claims 1 to 3, characterized in that: The step of obtaining the parameter vector of the weld includes: Obtaining the length, width and height of the weld; The parameter vector of the weld is constructed using the length, the width and the height as elements.

10. The method according to any one of claims 1 to 3, characterized in that: The vector distance includes Mahalanobis distance or Euclidean distance.

11. A welding quality evaluation device, characterized in that: include: A parameter vector acquisition module, used to acquire a parameter vector of a weld, wherein the parameter vector includes parameter values ​​of a plurality of welding characteristic parameters of the weld; A vector distance acquisition module, used to acquire the vector distance between the parameter vector and a historical sample vector set; each vector in the historical sample vector set includes historical sample values ​​of the multiple welding characteristic parameters; An evaluation module is used to evaluate whether the welding quality of the weld is abnormal based on the vector distance.

12. An electronic device, characterized in that: The device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the welding quality evaluation method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the welding quality evaluation method according to any one of claims 1 to 10.

Images