Screw tightening abnormity identification method, device and equipment and storage medium

Abstract

The invention discloses a screw tightening abnormity identification method and device, equipment and a storage medium, and belongs to the field of abnormity detection. The method comprises the following steps: acquiring tightening data of a to-be-tested screw in a tightening process; removing abnormal data from the tightening data, and determining to-be-tested data; performing feature extraction on the to-be-detected data to determine feature data; the feature data are input into an abnormity recognition model, the recognition result of the to-be-detected screw is obtained, the recognition result is used for representing whether the screwing process of the to-be-detected screw is abnormal or not, and the abnormity recognition model is obtained through training based on the feature data of the sample screw, the marking result corresponding to the feature data of the sample screw and the category of the sample screw. According to the method, the screwing process of the screw can be monitored on line through the abnormity recognition model, so that abnormity in the screwing process can be found in time, effective monitoring of the abnormity in the screwing process is achieved, and then the influence of screwing of the screw on the battery quality is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of anomaly detection, and in particular to a method, device, equipment and storage medium for identifying screw tightening anomalies. Background Art

[0002] Screws are one of the most commonly used parts in industry, and the tightening quality of each screw will directly affect the quality of the product. In the battery manufacturing industry, screws are also one of the important parts to complete the packaging of each battery pack. They are of great significance to ensure the normal charging and discharging of the battery, safe and reliable operation, etc. If there is a problem with the tightening quality of the screws, it may cause serious safety accidents.

[0003] The commonly used method for identifying screw tightening anomalies is to detect the tightening torque and tightening angle after the screw is tightened. However, this method will cause anomalies (such as screw skew, screw thread slippage, etc.) generated during the tightening process to be unable to be identified, thereby affecting product quality. Summary of the invention

[0004] The embodiments of the present application provide a method, device, equipment and storage medium for identifying screw tightening anomalies, so as to effectively monitor anomalies generated during the screw tightening process and reduce the impact of screw tightening on battery quality.

[0005] In order to solve the above technical problems, the embodiments of the present application disclose the following technical solutions:

[0006] In a first aspect, a method for identifying screw tightening anomalies is provided, the method comprising:

[0007] Obtaining the tightening data of the screw to be tested during the tightening process;

[0008] Removing abnormal data from the tightening data to determine the data to be tested;

[0009] Extracting features from the data to be tested to determine feature data;

[0010] The feature data is input into an abnormality recognition model to obtain an identification result of the screw to be tested, wherein the identification result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is trained based on the feature data of a sample screw, the marking result corresponding to the feature data of the sample screw, and the category of the sample screw.

[0011] In some embodiments, before obtaining the tightening data of the screw to be tested during the tightening process, the method further includes:

[0012] Building the architecture of the anomaly recognition model based on the CART decision tree;

[0013] Obtain tightening data of sample screws;

[0014] Determining the category of the sample screw based on the tightening data of the sample screw, where the category of the sample screw includes a negative sample screw and a positive sample screw;

[0015] Performing feature extraction and abnormal marking on the tightening data of the sample screw to obtain the feature data of the sample screw and the marking result corresponding to the feature data;

[0016] Based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the abnormality recognition model is trained and verified, and the trained abnormality recognition model is obtained.

[0017] In some embodiments, the tightening data includes torque data; and determining the category of the sample screw based on the tightening data of the sample screw includes:

[0018] Based on the torque data of the sample screw, determine the torque mean, torque maximum, torque distribution kurtosis, and torque distribution skewness corresponding to the sample screw;

[0019] Based on the triple standard deviation method, determining the first outlier sample screw having an abnormal torque mean value and / or an abnormal torque maximum value from all the sample screws;

[0020] Based on a cluster analysis method, determining a second outlier sample screw having abnormal torque distribution kurtosis and / or abnormal torque distribution skewness from all the sample screws;

[0021] The first outlier sample screw and the second outlier sample screw are determined as the negative sample screws, and the remaining sample screws except the first outlier sample screw and the second outlier sample screw among all the sample screws are determined as the positive sample screws.

[0022] In some embodiments, the tightening data includes torque data and angle data; feature extraction and abnormality marking are performed on the tightening data of the sample screw to obtain the feature data of the sample screw and the marking result corresponding to the feature data, including:

[0023] Extracting characteristic data of the sample screw based on the torque data and the angle data of the sample screw, wherein the characteristic data includes a plurality of characteristic values;

[0024] Each of the characteristic values ​​is marked as abnormal, and a marking result corresponding to each of the characteristic values ​​is determined, wherein the marking result includes normal and abnormal.

[0025] In some embodiments, the tightening process includes a screw-in stage and a tightening stage, and the plurality of characteristic values ​​include a maximum torque value of the tightening process, a tightening angle difference value of the tightening process, a climbing angle value of the tightening process, a maximum torque value of the screw-in stage, a torque mean value of the screw-in stage, and a torque distribution kurtosis of the tightening stage;

[0026] The step of marking each of the characteristic values ​​as abnormal and determining a marking result corresponding to each of the characteristic values ​​includes:

[0027] Based on the comparison results of the maximum torque value of the tightening process, the tightening angle difference value of the tightening process, and the climbing angle value of the tightening process with the corresponding preset thresholds, determine the marking result corresponding to each of the characteristic values ​​of the maximum torque value of the tightening process, the tightening angle difference value of the tightening process, and the climbing angle value of the tightening process;

[0028] Based on the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage corresponding to all the sample screws, determine the marking result corresponding to each of the characteristic values ​​in the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage by triple standard deviation method;

[0029] Based on the torque distribution kurtosis of all the tightening stages, a labeling result corresponding to each characteristic value in the torque distribution kurtosis of all the tightening stages is determined by a cluster analysis method.

[0030] In some embodiments, before training and verifying the abnormality recognition model based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the method further includes:

[0031] Based on the feature data of the negative sample screws and the marking results corresponding to the feature data, the negative sample screws are oversampled to obtain feature data of newly added negative sample screws and the marking results corresponding to the feature data, wherein the deviation between the sum of the number of the negative sample screws and the newly added negative sample screws and the number of the positive sample screws meets a preset condition.

[0032] In some embodiments, before determining the category of the sample screw based on the tightening data of the sample screw, the method further includes:

[0033] The tightening data of the sample screws are cleaned to remove abnormal samples from the sample screws.

[0034] In some embodiments, the feature data includes a plurality of feature values; and the method further includes:

[0035] If the recognition result indicates that there is an abnormality in the tightening process of the screw to be tested, obtaining decision path information of the feature data in the abnormality recognition model;

[0036] Based on the decision path information, feature values ​​of the feature data for which a marking result is abnormal are determined.

[0037] In a second aspect, a device for identifying abnormal screw tightening is provided, the device comprising:

[0038] A data acquisition module is used to obtain the tightening data of the screw to be tested during the tightening process;

[0039] A data cleaning module, used to remove abnormal data from the tightening data and determine the data to be tested;

[0040] A feature extraction module is used to extract features from the data to be tested and determine feature data;

[0041] The abnormality recognition module is used to input the characteristic data into an abnormality recognition model to obtain the recognition result of the screw to be tested, and the recognition result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is obtained based on the characteristic data of the sample screw, the marking result corresponding to the characteristic data of the sample screw and the category training of the sample screw.

[0042] In a third aspect, an electronic device is provided, the electronic device comprising a memory and a processor;

[0043] The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method according to any one of the first aspects is performed.

[0044] According to a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of the first aspects is executed.

[0045] One of the above technical solutions has the following advantages or beneficial effects:

[0046] Compared with the prior art, the present invention provides a method for identifying screw tightening anomalies, including: obtaining tightening data of the screw to be tested during the tightening process; removing abnormal data from the tightening data to determine the data to be tested; extracting features from the data to be tested to determine feature data; inputting the feature data into an abnormal recognition model to obtain an identification result of the screw to be tested, the identification result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested, and the abnormal recognition model is obtained based on the feature data of the sample screw, the labeling result corresponding to the feature data of the sample screw, and the category training of the sample screw. The method provided in the present application can monitor the tightening process of the screw online through the abnormal recognition model, so that the anomaly in the tightening process can be discovered in time, and the effective monitoring of the abnormality in the tightening process can be achieved, thereby reducing the impact of screw tightening on battery quality.

[0047] A screw tightening abnormality identification device of the present application can monitor the screw tightening process online, so that abnormalities in the tightening process can be discovered in time, and effective monitoring of abnormalities in the tightening process can be achieved, thereby reducing the impact of screw tightening on battery quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 A schematic diagram of the overall process of the method for obtaining an abnormality recognition model provided in an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the structure of the CART decision tree provided in the embodiment of the present application;

[0051] Figure 3 A schematic diagram of a torque distribution curve of a sample screw provided in an embodiment of the present application;

[0052] Figure 4 A schematic diagram of the angle distribution curve of a sample screw provided in an embodiment of the present application;

[0053] Figure 5 A schematic diagram of the overall process of a method for identifying abnormal screw tightening provided in an embodiment of the present application;

[0054] Figure 6 A schematic diagram of a specific process of a method for identifying abnormal screw tightening provided in an embodiment of the present application;

[0055] Figure 7 This is a schematic diagram of the structure of a device for identifying abnormal screw tightening according to an embodiment of the present application;

[0056] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0057] Reference numerals:

[0058] 701-data acquisition module; 702-data cleaning module; 703-feature extraction module; 704-abnormal identification module; 801-memory; 802-processor. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0060] In the description of the present application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying 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" and "second" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the present application, "multiple" means two or more, and at least one means one, two or more, unless otherwise clearly and specifically defined.

[0061] Since the tightening quality of the screws will have a great impact on the quality of the battery, the tightening quality of the screws is usually tested in battery production workshops. The commonly used method for identifying screw tightening anomalies is to detect the tightening torque and tightening angle after the screws are tightened. For example, it detects whether the tightening torque reaches the preset torque value of 6N, or detects whether the tightening angle exceeds 62 degrees. However, this method of judging anomalies through a single result data not only monitors too little data during the tightening process and is relatively monotonous, making it impossible to identify some anomalies in the tightening process (such as screw skew, screw slippage, and other curve jitters), but also relies solely on a single threshold to judge anomalies. It is easy to leave safety hazards, product quality cannot be guaranteed, and manual re-inspection consumes manpower and cannot identify most anomalies.

[0062] In view of this, an embodiment of the present application provides a method for identifying screw tightening anomalies, which obtains the tightening data of the screw tightening process, extracts feature data therefrom, and finally puts it into a pre-built and trained anomaly recognition model for anomaly recognition to determine whether there is an abnormality in the tightening process of the screw to be tested, thereby timely discovering anomalies in the tightening process, achieving effective monitoring of abnormalities in the tightening process, and thereby reducing the impact of screw tightening on battery quality.

[0063] The following first introduces the process of acquiring the anomaly recognition model in the embodiment of the present application.

[0064] See also Figure 1 , Figure 1 The overall flow diagram of the method for obtaining an abnormality recognition model provided in the embodiment of the present application. Before judging the abnormality of the tightening process of the screw to be tested, the abnormality recognition model can be obtained by the following steps:

[0065] Step 101: Build an architecture of anomaly recognition model based on CART decision tree.

[0066] Specifically, CART (Classification and Regression Trees) decision tree is a machine learning model based on decision trees, which is used to solve classification and regression problems. CART decision tree can select one feature at a time according to the input sample data for segmentation, generate two child nodes, and continue to segment until the sample category can be classified.

[0067] See also Figure 2 , Figure 2 A schematic diagram of the structure of the CART decision tree provided in the embodiment of the present application. The parameters of the CART decision tree can be set according to the actual situation. For example, the embodiment of the present application can set the following parameters for the CART decision tree: use Gini impurity to select the optimal splitting feature; set the tree depth max_depth to 3 to 10; set the splitter strategy to determine how the decision tree selects the best classification point; set the max_features parameter to control the maximum number of features considered by the decision tree when splitting a node. Figure 2 As shown in the figure, the tree depth of the CART decision tree is 4. Each tree node has x[i], where i indicates which feature the node uses for classification and the value of x[i] is the classification threshold; gini refers to the Gini coefficient of the node; samples refers to the number of samples to be classified by the node; and value[0,82] in which 0 represents the classification category of the node and 82 represents that there are 82 samples belonging to this category.

[0068] Step 102: Obtain tightening data of a sample screw.

[0069] Specifically, the sample screws may be the screws tightened on the production line within a preset historical period. Each screw on the production line may be tightened by a screw gun within the preset historical period. The tightening data may be collected by a screw gun. Exemplarily, the acquisition frequency may be set to 2ms (milliseconds). The tightening process of each screw may collect data of 1000 sample points, which may include characteristic variables such as torque, angle, and current. In other words, the tightening data of the sample screw may include at least one of torque data, angle data, and current data.

[0070] In some embodiments, after executing step 102 and before executing step 103, the method of the embodiment of the present application may further include:

[0071] The tightening data of the sample screws are cleaned to remove abnormal samples from the sample screws.

[0072] Specifically, since the screws will be recognized before being tightened, and the recognized data will interfere with the subsequent feature extraction, the recognized data and blank data need to be cleaned up in the data cleaning stage.

[0073] Exemplarily, when performing data cleaning, first, the integrity of the tightening data of each sample screw can be checked. For example, each sample screw collects 1000 data points during the tightening process, then the sample screws whose data missing does not meet 1000 sample points, and the sample screws containing a large number of abnormal characters or missing characters are determined as abnormal samples. Then, traverse the angle data and torque data of each remaining sample screw. If the maximum value of the angle is less than or equal to 2000° (degrees) and the maximum value of the torque is less than or equal to 1N (Newton), then the sample screw is determined to be a cap-recognition sample, and the cap-recognition sample is also determined as an abnormal sample. Finally, remove the abnormal samples from all sample screws, and save the remaining sample screws for subsequent model training.

[0074] Through the above solution, abnormal samples can be cleaned up to ensure that the data used for training can fit most tightening scenarios, which is conducive to improving the accuracy of subsequent training.

[0075] Step 103: Based on the tightening data of the sample screws, determine the category of the sample screws, where the category of the sample screws includes negative sample screws and positive sample screws.

[0076] Specifically, based on the tightening data of the sample screws, the sample screws are classified into positive sample screws and negative sample screws by means of data mining. The positive sample screws refer to the sample screws with normal tightening process, and the negative sample screws refer to the sample screws with abnormal tightening process.

[0077] In some embodiments, step 103 may be implemented by the following steps:

[0078] Step 1: Based on the torque data of the sample screws, determine the torque mean, torque maximum, torque distribution kurtosis, and torque distribution skewness corresponding to the sample screws.

[0079] Specifically, the torque data is the rotational moment data applied to the screw. The torque data of the sample screw can be visualized through a line graph to determine the overall change trend of the torque through the torque change curve.

[0080] Among them, the torque mean refers to the average torque value of the sample screw during the entire tightening process. The torque maximum includes the maximum torque value and the minimum torque value, which respectively refer to the maximum torque value and the minimum torque value of the sample screw during the entire tightening process. The torque distribution kurtosis refers to the sharpness of the torque distribution morphology of the sample screw during the entire tightening process. The torque distribution skewness refers to the degree of deflection or asymmetry of the torque distribution of the sample screw during the entire tightening process, which is used to describe the degree of deviation of the torque curve from symmetry. Both the torque distribution kurtosis and the torque distribution skewness can be calculated based on the torque value, standard deviation, torque mean and number of data points of each data point, and the embodiments of the present application will not be repeated here.

[0081] It can be understood that each sample screw corresponds to a torque mean, a torque maximum, a torque distribution kurtosis and a torque distribution skewness.

[0082] Step 2: Based on the triple standard deviation method, determine the first outlier sample screw with abnormal torque mean and / or abnormal torque maximum from all sample screws.

[0083] Specifically, the first outlier sample screw refers to a sample screw with an abnormal torque mean value and / or an abnormal maximum torque value and / or an abnormal minimum torque value. The triple standard deviation method is mean+3σ, where mean is the mean value of the data set and σ is the standard deviation of the data set.

[0084] Exemplarily, based on the torque mean values ​​corresponding to all sample screws, the average value mean1 of the torque mean values ​​and the standard deviation σ1 of the torque mean values ​​are determined, and the torque mean values ​​that exceed the range of mean1±3σ1 are determined as torque mean anomalies, and the sample screws corresponding to the torque mean values ​​are determined as the first outlier sample screws. Based on the torque maximum values ​​corresponding to all sample screws, the average value mean2 of the torque maximum values ​​and the standard deviation σ2 of the torque maximum values ​​are determined, and the torque maximum values ​​that exceed the range of mean2±3σ2 are determined as torque maximum anomalies, and the sample screws corresponding to the torque maximum values ​​are also determined as the first outlier sample screws.

[0085] Step three, based on the cluster analysis method, determine the second outlier sample screw with abnormal torque distribution kurtosis and / or abnormal torque distribution skewness from all the sample screws.

[0086] Specifically, the second outlier sample screw refers to a sample screw with abnormal torque distribution kurtosis and / or abnormal torque distribution skewness. The cluster analysis method may adopt a KMeans cluster analysis method.

[0087] Exemplarily, the kurtosis of the torque distribution corresponding to all sample screws forms a kurtosis sample space, and the skewness of the torque distribution corresponding to all sample screws forms a skewness sample space. Randomly select k points as cluster center points in the kurtosis sample space and the skewness sample space respectively, calculate the distance from each point to the cluster center, assign the point to the nearest cluster center, and calculate the average value of the cluster sample point for each cluster, move the cluster center to the average value, and then randomly select k points as cluster center points again, and repeat the above steps until the cluster center in each sample space no longer changes significantly, that is, the best clustering result is achieved. Calculate the distance from each sample point to the cluster center, set the cluster radius threshold, and determine the sample screws corresponding to the discrete points outside the cluster radius threshold as the second outlier sample screws.

[0088] It is understandable that the optimal k value can be calculated based on the kurtosis and skewness. The initial k value can be defined as 3 to 10, and the k value calculated each time is not necessarily the same. By visually observing the classification effect, the cluster center can be regarded as a sample circle. For example, in a certain training, the cluster radius threshold is set to 1.412, that is, normal samples are divided into a circle with a cluster radius of 1.412, and those outside 1.412 are abnormal samples. It can be determined according to the actual situation. The embodiment of the present application does not specifically limit the selection of the k value and the setting of the cluster radius threshold.

[0089] Step 4: determine the first outlier sample screw and the second outlier sample screw as negative sample screws, and determine the remaining sample screws in all sample screws except the first outlier sample screw and the second outlier sample screw as positive sample screws.

[0090] In some embodiments, after executing step 4, the negative sample screws determined by steps 1 to 4 can also be compared with the abnormal samples of process verification to determine whether the marking result of the negative sample screw is wrong. When the marking result of the negative sample screw is wrong, it can be corrected according to the abnormal samples of process verification to improve the accuracy of training sample construction.

[0091] Through the above scheme, abnormal samples are marked from the sample screws using a statistical method, which can avoid the problem of safety hazards that may be easily left by relying solely on a single threshold to judge the difference. As a result, abnormal samples can be distinguished more reasonably, and the over-kill rate and missed kill rate are less likely to occur. The results are also less likely to be affected by the equipment and have better versatility.

[0092] Step 104: extract features and mark abnormalities on the tightening data of the sample screws to obtain feature data of the sample screws and marking results corresponding to the feature data.

[0093] Specifically, feature data is extracted from the tightening data, and each feature value in the feature data is marked as normal or abnormal.

[0094] In some embodiments, step 104 may be implemented by the following steps:

[0095] Step 1: extracting characteristic data of the sample screw based on the torque data and angle data of the sample screw, where the characteristic data includes a plurality of characteristic values.

[0096] Specifically, the angle data refers to the angle of rotation of the screw. The angle data of the sample screw can be visualized through a line graph to determine the overall change trend of the angle through the angle change curve.

[0097] The selection of the characteristic value can be performed by comparing the torque curve and the angle curve of the normal sample with the abnormal sample in advance to select the required classification features. It can be understood that the performance of the selected characteristic value on the normal sample and the abnormal sample is different.

[0098] Please also read Figure 3 and Figure 4 , Figure 3 A schematic diagram of the torque distribution curve of a sample screw provided in an embodiment of the present application, Figure 4 Schematic diagram of the angle distribution curve of the sample screw provided for the embodiment of the present application. Wherein, the horizontal axis represents the column index of the data point, for example, the horizontal axis is 100, indicating that the sample collects the 100th data as the time series. Exemplarily, the screw can be tightened in two steps, and the tightening process includes a screw-in stage and a tightening stage. In the screw-in stage, the torque reaches a first preset value T1, and in the tightening stage, the torque reaches a second preset value T2. Due to the different tightening time of each screw, the sample may have redundant data, but it has no effect on the result, and this situation can be ignored in the embodiment of the present application. Multiple characteristic values ​​may include the maximum torque value of the tightening process, the tightening angle difference of the tightening process, the climbing angle value of the tightening process, the maximum torque value of the screw-in stage, the torque mean of the screw-in stage, and the torque distribution kurtosis of the tightening stage.

[0099] Specifically, the maximum torque value of the tightening process can be obtained by traversing the torque data of each sample screw during the tightening process. The tightening angle difference during the tightening process can represent how many turns the tightening gun has made during the tightening process (one turn is 360°). The index x0 of the start screwing point can be obtained by reading the torque curve of the sample screw, and then the angle value a0 corresponding to x0 is obtained. The maximum value of the tightening angle is set to a max, that is, the tightening angle difference a of the entire tightening process can be obtained diff =a max -a0. The climbing angle value of the tightening process can be obtained by reading the torque curve of the sample screw, obtaining the index x5 corresponding to the torque reaching 5N, and then obtaining the angle value a5 corresponding to x5, that is, the climbing angle value a can be obtained. max -a5.

[0100] Since there are many differences between the first part (i.e., the screw-in stage) and the second part (i.e., the tightening stage) of the torque curve of the two-step tightening, the embodiment of the present application extracts part of the torque curve as a feature. First, take the index of the return to zero point after the first peak as the dividing point between the screw-in stage and the tightening stage. The reason for the return to zero point after the first peak is that the sample screw is tightened in two steps, that is, the screw is first tightened to the surface of the component, which is equivalent to the screw-in stage. The components and bolts will not undergo elastic deformation. Then, in the tightening stage, the connector will produce axial loads that cause elastic deformation. The deformed bolts will be subjected to impact loads. Therefore, it is appropriate to loosen a certain angle, that is, return to zero after the first peak, and then tighten to the target value. This can reduce torque loss. In some examples, a counter can be initialized to search forward from index x5 until a sample point x is found. i The corresponding torque value is closest to 0, that is, the index spilt_index of the demarcation point is obtained. In other examples, the index spilt_index of the demarcation point can also be obtained by other methods, which is not specifically limited in the embodiment of the present application.

[0101] The maximum torque value in the screw-in stage refers to the maximum torque value of the torque curve before the index spilt_index of the dividing point, that is, the first peak value. The mean torque value in the screw-in stage refers to the mean torque value of the torque curve before the index spilt_index of the dividing point. The kurtosis of the torque distribution in the tightening stage refers to the steepness of the torque curve after the index spilt_index of the dividing point.

[0102] It is understandable that the index of the embodiment of the present application is used to characterize a sample point in the tightening process of the sample screw. In addition, the multiple characteristic values ​​may also include other characteristic values, such as the maximum torque value in the tightening stage, that is, the maximum torque value of the torque curve after the index spilt_index of the dividing point, that is, the second peak value, which is not specifically limited in the embodiment of the present application.

[0103] Step 2: Mark each feature value as abnormal and determine the marking result corresponding to each feature value. The marking results include normal and abnormal.

[0104] In some embodiments, the marking result corresponding to each feature value may be determined in the following manner:

[0105] Based on the comparison results of the maximum torque value of the tightening process, the tightening angle difference of the tightening process, and the climbing angle value of the tightening process with the corresponding preset thresholds, the marking result corresponding to each characteristic value of the maximum torque value of the tightening process, the tightening angle difference of the tightening process, and the climbing angle value of the tightening process is determined.

[0106] Based on the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage corresponding to all sample screws, the marking result corresponding to each characteristic value in the maximum torque value of all screwing-in stages and the mean torque value of the screwing-in stage is determined by the triple standard deviation method.

[0107] Based on the torque distribution kurtosis of all tightening stages, the labeling result corresponding to each eigenvalue in the torque distribution kurtosis of all tightening stages is determined by cluster analysis method.

[0108] In some examples, the preset threshold corresponding to the maximum torque value of the tightening process may include a first threshold and a second threshold greater than the first threshold, the first threshold may be set to the second preset value T2-0.5N, and the second threshold may be set to the second preset value T2+0.5N. In other words, if the maximum torque value of the tightening process of the sample screw is less than T2-0.5N or the maximum torque value of the tightening process is greater than T2+0.5N, the maximum torque value of the tightening process of the sample screw is marked as an abnormal feature; if the maximum torque value of the tightening process of the sample screw is greater than or equal to T2-0.5N and less than or equal to T2+0.5N, the maximum torque value of the tightening process of the sample screw is marked as a normal feature.

[0109] The preset threshold corresponding to the tightening angle difference of the tightening process may include a third threshold. Exemplarily, the normal tightening angle difference is about 3000°, and the third threshold may be set to 2000°. In other words, if the tightening angle difference of the tightening process is less than 2000°, it indicates that the sample screw is not tightened enough times, and the tightening angle difference of the tightening process of the sample screw is marked as an abnormal feature; if the tightening angle difference of the tightening process is greater than or equal to 2000°, the tightening angle difference of the tightening process of the sample screw is marked as a normal feature.

[0110] The preset threshold corresponding to the climbing angle value of the tightening process may include a fourth threshold. Exemplarily, the fourth threshold can be set to 62°. In other words, if the climbing angle value of the tightening process is greater than 62°, the climbing angle value of the tightening process of the sample screw is marked as an abnormal feature; if the climbing angle value of the tightening process is less than or equal to 62°, the climbing angle value of the tightening process of the sample screw is marked as a normal feature. In this way, since 62° is the tightening angle threshold determined by the process, the torque reaches the tightening threshold after reaching 5N, and it is tightened with a slight twist (the tightening angle threshold of 62° at this stage). Tightening more (angle greater than 62°) will result in excessive tightening force, which may damage the screw or thread. Therefore, setting the fourth threshold to 62° can effectively improve the problem of damage to the screw or thread.

[0111] According to the maximum torque values ​​of the screwing-in stage corresponding to all sample screws, the mean value of the maximum torque value mean3 and the standard deviation of the maximum torque value σ3 are determined, and the maximum torque value exceeding the range of mean3±3σ3 is determined as an abnormal feature, and the maximum torque value within the range of mean3±3σ3 is determined as a normal feature. According to the torque mean values ​​of the screwing-in stage corresponding to all sample screws, the mean value of the torque mean mean4 and the standard deviation of the torque mean σ4 are determined, and the torque mean value exceeding the range of mean4±3σ4 is determined as an abnormal feature, and the torque mean value within the range of mean4±3σ4 is determined as a normal feature.

[0112] Through the above scheme, since the period before the dividing point spilt_index is from the first stage of screwing in to the completion of the first peak, the torque curve in this stage is relatively stable except for the first peak. The mean analysis of the curve before the dividing point split_index can effectively detect the stability of the curve in this stage.

[0113] According to the kurtosis sample space formed by the kurtosis of the torque distribution in all tightening stages, k points are randomly selected as cluster center points in the kurtosis sample space, the distance from each point to the cluster center is calculated, the point is assigned to the nearest cluster center, and the average value of the cluster sample points is calculated for each cluster, the cluster center is moved to the average value, and then k points are randomly selected again as cluster center points, and the above steps are repeated until the cluster center in each sample space no longer changes significantly, that is, the best clustering result is achieved. The distance from each sample point to the cluster center is calculated, and after setting the cluster radius threshold, the torque distribution kurtosis corresponding to the discrete points outside the cluster radius threshold is determined as an abnormal feature, and the torque distribution kurtosis corresponding to the discrete points within the cluster radius threshold is determined as a normal feature.

[0114] Through the above scheme, since the curve after the index spilt_index of the dividing point is an important stage of bolt tightening, that is, the second peak, the curve fluctuates greatly, and the kurtosis is used to measure the sharpness and peak state of this section of the curve, which can well measure the curve in this stage, thereby making the feature extraction more accurate.

[0115] In some embodiments, after executing step 104 and before executing step 105, the method of the embodiment of the present application may further include the following steps:

[0116] Based on the feature data of the negative sample screws and the marking results corresponding to the feature data, the negative sample screws are oversampled to obtain the feature data of the newly added negative sample screws and the marking results corresponding to the feature data, wherein the deviation between the sum of the number of negative sample screws and the newly added negative sample screws and the number of positive sample screws meets a preset condition.

[0117] Specifically, since the number of negative sample screws is usually smaller than the number of positive sample screws, for example, 3000 sample screws are extracted, of which there are only 200 negative sample screws, the negative sample screws need to be oversampled so that the number of positive and negative samples can be balanced.

[0118] In some examples, the SMOTE method can be used to oversample negative sample screws, analyze the K neighboring samples of each negative sample screw, and synthesize new negative sample screws by randomly interpolating between the sample points of each negative sample screw and the sample points of its adjacent negative sample screws, thereby increasing the number of negative samples to 10 times the original number, making the number of positive and negative samples more balanced. In this way, the SMOTE method can focus on minority samples, generate new samples by interpolation to balance the ratio of positive and negative samples, and thus pay more attention to the distribution and characteristics of minority samples.

[0119] In other examples, undersampling, data augmentation and other methods may also be used to generate new negative sample screws, and the embodiments of the present application do not specifically limit this.

[0120] The above solution can avoid the situation where the number of abnormal samples is small, which easily leads to model overfitting during training, and improve the accuracy of subsequent model training.

[0121] Step 105: Based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the abnormality recognition model is trained and verified, and the trained abnormality recognition model is obtained.

[0122] In some examples, the anomaly recognition model can be trained and evaluated multiple times using grid search and ten-fold cross validation to find the best parameter combination. Figure 2Take the CART decision tree shown in the figure as an example. After training and verification, the tree depth max_depth is 3, the splitter strategy is best, and the max_features parameter is 3. After completing training and verification, the test set can be input into the trained model for testing, and the confusion matrix and classification report can be used to observe the classification accuracy of the model. Finally, you can print out Figure 2 The decision tree shown is easy to observe, Figure 2 As shown in the figure, the decision tree has classification features for each node, as well as classification thresholds, Gini coefficients, and sample sizes for classification. By tracing the path of feature values ​​in the decision tree, we can further learn the specific abnormal features in the abnormal samples, such as Figure 2 In the example, x[7]<=2.737 is one of the classification thresholds. Finally, the feature and the abnormality type can be output, such as "the second peak is smaller".

[0123] Through the above scheme, machine learning is used to train the anomaly recognition model. The training sample data is the feature data of the extracted sample screw tightening process and the anomaly marking result of each feature value, which can improve the anomaly recognition rate of the model. In addition, the training accuracy can be further improved by balancing the number of positive and negative samples. Through the anomaly recognition model, the screw tightening quality can be monitored online, and an alarm can be issued in time when an abnormality is found in the tightening data, reducing the impact of screw tightening on battery quality.

[0124] The following introduces the method of detecting anomalies in the screw tightening process through an anomaly recognition model in an embodiment of the present application.

[0125] See also Figure 5 , Figure 5 The overall flow chart of a method for identifying abnormal screw tightening provided in an embodiment of the present application is as follows:

[0126] Step 501: Obtain tightening data of the screw to be tested during the tightening process.

[0127] Specifically, the screws to be tested may be the screws currently being tightened on the production line. Data of multiple sample points of the screws to be tested during the tightening process may be collected, and the tightening data of the screws to be tested may be at least one of torque data, angle data, and current data.

[0128] Step 502: Remove abnormal data from the tightening data and determine the data to be tested.

[0129] Exemplarily, the abnormal data may include capped data, empty data, and garbled data. In this way, removing the abnormal data from the tightening data can remove data that is not helpful for feature extraction and improve the accuracy of subsequent predictions.

[0130] Step 503: extract features from the data to be tested and determine feature data.

[0131] Exemplarily, characteristic data of the screw to be tested can be extracted based on the torque data and angle data of the screw to be tested. The characteristic data of the screw to be tested includes multiple characteristic values, and the multiple characteristic values ​​can include the maximum torque value of the tightening process, the tightening angle difference of the tightening process, the climbing angle value of the tightening process, the maximum torque value of the screwing-in stage, the torque mean of the screwing-in stage, and the torque distribution kurtosis of the tightening stage.

[0132] It should be noted that the specific feature extraction process can refer to the feature extraction process in the aforementioned embodiment, and the embodiments of the present application will not be repeated here.

[0133] Step 504: Input the feature data into the abnormality recognition model to obtain the recognition result of the screw to be tested. The recognition result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is trained based on the feature data of the sample screw, the marking result corresponding to the feature data of the sample screw, and the category of the sample screw.

[0134] That is to say, the extracted feature data is input into the abnormality recognition model, and the abnormality recognition model can output the result of whether there is an abnormality in the tightening process of the screw to be tested.

[0135] See also Figure 6 , Figure 6 A schematic diagram of a specific process of a method for identifying screw tightening anomalies provided in an embodiment of the present application. In some embodiments, after executing step 504, the method of the embodiment of the present application may further include the following steps:

[0136] Step 505: If the recognition result indicates that there is an abnormality in the tightening process of the screw to be tested, then the decision path information of the feature data in the abnormality recognition model is obtained.

[0137] Step 506: Based on the decision path information, determine the feature value of the feature data for which the result is marked as abnormal.

[0138] It can be understood that since the anomaly identification model adopts the CART decision tree, the node of each feature value and the classification threshold can be obtained from the decision tree. Then, after the tightening process of the screw to be tested is judged to be abnormal, the feature value marked as abnormal in the feature data can be obtained from the current decision path information of the model.

[0139] It can be understood that the method of the embodiment of the present application extracts multi-dimensional features through data mining, improves the anomaly detection rate, can output the abnormal type for the abnormal tightening curve, effectively reduces the time and cost of manual re-inspection, and in addition, by using the abnormal recognition model to monitor the tightening process of the screws online, it is possible to promptly detect anomalies in the tightening process, achieve effective monitoring of the anomalies in the tightening process, and thereby reduce the impact of screw tightening on battery quality.

[0140] Accordingly, see Figure 7 , Figure 7 The structure diagram of the screw tightening anomaly identification device according to the embodiment of the present application is as follows: The screw tightening anomaly identification device according to the embodiment of the present application comprises a data acquisition module 701 , a data cleaning module 702 , a feature extraction module 703 and an anomaly identification module 704 .

[0141] The data acquisition module 701 is used to obtain the tightening data of the screw to be tested during the tightening process.

[0142] The data cleaning module 702 is used to remove abnormal data from the tightening data and determine the data to be tested.

[0143] The feature extraction module 703 is used to extract features from the data to be tested and determine feature data.

[0144] The abnormality recognition module 704 is used to input the characteristic data into the abnormality recognition model to obtain the recognition result of the screw to be tested. The recognition result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is obtained based on the characteristic data of the sample screw, the marking result corresponding to the characteristic data of the sample screw and the category training of the sample screw.

[0145] In some embodiments, the device further includes a model training module, which is used to:

[0146] The architecture of anomaly recognition model based on CART decision tree.

[0147] Obtain tightening data for a sample screw.

[0148] Based on the tightening data of the sample screws, the category of the sample screws is determined, and the category of the sample screws includes negative sample screws and positive sample screws.

[0149] Feature extraction and abnormal marking are performed on the tightening data of the sample screws to obtain the feature data of the sample screws and the marking results corresponding to the feature data.

[0150] Based on the feature data of the sample screws, the labeling results corresponding to the feature data, and the category of the sample screws, the abnormality recognition model is trained and verified, and the trained abnormality recognition model is obtained.

[0151] In some embodiments, the tightening data includes torque data. The model training module is specifically used for:

[0152] Based on the torque data of the sample screws, the torque mean, torque maximum, torque distribution kurtosis, and torque distribution skewness corresponding to the sample screws are determined.

[0153] Based on the three-times standard deviation method, the first outlier sample screw with abnormal torque mean value and / or abnormal torque maximum value is determined from all sample screws.

[0154] Based on the cluster analysis method, a second outlier sample screw with abnormal torque distribution kurtosis and / or abnormal torque distribution skewness is determined from all the sample screws.

[0155] The first outlier sample screw and the second outlier sample screw are determined as negative sample screws, and the remaining sample screws except the first outlier sample screw and the second outlier sample screw among all the sample screws are determined as positive sample screws.

[0156] In some embodiments, the tightening data includes torque data and angle data. The model training module is specifically used for:

[0157] Based on the torque data and angle data of the sample screw, feature data of the sample screw is extracted, and the feature data includes a plurality of feature values.

[0158] Each feature value is marked as abnormal, and the marking result corresponding to each feature value is determined, and the marking result includes normal and abnormal.

[0159] In some embodiments, the tightening process includes a screw-in stage and a tightening stage, and the multiple characteristic values ​​include the maximum torque value of the tightening process, the tightening angle difference of the tightening process, the climbing angle value of the tightening process, the maximum torque value of the screw-in stage, the torque mean of the screw-in stage, and the torque distribution kurtosis of the tightening stage.

[0160] The model training module is specifically used for:

[0161] Based on the comparison results of the maximum torque value of the tightening process, the tightening angle difference of the tightening process, and the climbing angle value of the tightening process with the corresponding preset thresholds, the marking result corresponding to each characteristic value of the maximum torque value of the tightening process, the tightening angle difference of the tightening process, and the climbing angle value of the tightening process is determined.

[0162] Based on the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage corresponding to all sample screws, the marking result corresponding to each characteristic value in the maximum torque value of all screwing-in stages and the mean torque value of the screwing-in stage is determined by the triple standard deviation method.

[0163] Based on the torque distribution kurtosis of all tightening stages, the labeling result corresponding to each eigenvalue in the torque distribution kurtosis of all tightening stages is determined by cluster analysis method.

[0164] In some embodiments, before training and verifying the abnormality recognition model based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the model training module is further used to:

[0165] Based on the feature data of the negative sample screws and the marking results corresponding to the feature data, the negative sample screws are oversampled to obtain the feature data of the newly added negative sample screws and the marking results corresponding to the feature data, wherein the deviation between the sum of the number of negative sample screws and the newly added negative sample screws and the number of positive sample screws meets a preset condition.

[0166] In some embodiments, before determining the category of the sample screw based on the tightening data of the sample screw, the model training module is further configured to:

[0167] The tightening data of the sample screws are cleaned to remove abnormal samples from the sample screws.

[0168] In some embodiments, the feature data includes multiple feature values. The abnormality identification module 704 is also used to:

[0169] If the recognition result indicates that there is an abnormality in the tightening process of the screw to be tested, the decision path information of the feature data in the abnormality recognition model is obtained.

[0170] Based on the decision path information, feature values ​​in the feature data whose marking results are abnormal are determined.

[0171] It can be understood that the screw tightening abnormality identification device of the embodiment of the present application can monitor the screw tightening process online, so that abnormalities in the tightening process can be discovered in time, and effective monitoring of abnormalities in the tightening process can be achieved, thereby reducing the impact of screw tightening on battery quality.

[0172] Accordingly, see Figure 8 , Figure 8 The present application also provides an electronic device, including a memory 801 and a processor 802, wherein the memory 801 stores a computer program executable by the processor 802, and when the computer program is executed by the processor 802, the steps of the screw tightening abnormality identification method in the above-mentioned embodiment are performed.

[0173] It can be understood that the electronic device of the embodiment of the present application can monitor the tightening process of the screws online, so that abnormalities in the tightening process can be discovered in time, and effective monitoring of abnormalities in the tightening process can be achieved, thereby reducing the impact of screw tightening on battery quality.

[0174] Correspondingly, an embodiment of the present application further provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a processor, the steps of the screw tightening abnormality identification method in the aforementioned embodiment are executed.

[0175] It can be understood that the readable storage medium of the embodiment of the present application enables the processor to monitor the tightening process of the screws online, so that abnormalities in the tightening process can be discovered in time, and effective monitoring of abnormalities in the tightening process can be achieved, thereby reducing the impact of screw tightening on battery quality.

[0176] The above is a detailed introduction to a method, device, equipment and storage medium for identifying abnormal screw tightening provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the technical solution and its core idea of ​​the present application. Ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents; and these modifications or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiments of the present application.

Claims

1. A method for identifying abnormal screw tightening, characterized in that: The method comprises: Obtaining the tightening data of the screw to be tested during the tightening process; Removing abnormal data from the tightening data to determine the data to be tested; Extracting features from the data to be tested to determine feature data; The feature data is input into an abnormality recognition model to obtain an identification result of the screw to be tested, wherein the identification result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is trained based on the feature data of a sample screw, the marking result corresponding to the feature data of the sample screw, and the category of the sample screw.

2. The method for identifying abnormal screw tightening according to claim 1, characterized in that: Before obtaining the tightening data of the screw to be tested during the tightening process, the method further includes: Building the architecture of the anomaly recognition model based on the CART decision tree; Obtain tightening data of sample screws; Determining the category of the sample screw based on the tightening data of the sample screw, where the category of the sample screw includes a negative sample screw and a positive sample screw; Performing feature extraction and abnormal marking on the tightening data of the sample screw to obtain the feature data of the sample screw and the marking result corresponding to the feature data; Based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the abnormality recognition model is trained and verified, and the trained abnormality recognition model is obtained.

3. The method for identifying abnormal screw tightening according to claim 2, characterized in that: The tightening data includes torque data; and determining the type of the sample screw based on the tightening data of the sample screw includes: Based on the torque data of the sample screw, determine the torque mean, torque maximum, torque distribution kurtosis, and torque distribution skewness corresponding to the sample screw; Based on the triple standard deviation method, determining the first outlier sample screw having an abnormal torque mean value and / or an abnormal torque maximum value from all the sample screws; Based on a cluster analysis method, determining a second outlier sample screw having abnormal torque distribution kurtosis and / or abnormal torque distribution skewness from all the sample screws; The first outlier sample screw and the second outlier sample screw are determined as the negative sample screws, and the remaining sample screws except the first outlier sample screw and the second outlier sample screw among all the sample screws are determined as the positive sample screws.

4. The method for identifying abnormal screw tightening according to claim 2, characterized in that: The tightening data includes torque data and angle data; feature extraction and abnormal marking are performed on the tightening data of the sample screw to obtain the feature data of the sample screw and the marking result corresponding to the feature data, including: Extracting characteristic data of the sample screw based on the torque data and the angle data of the sample screw, wherein the characteristic data includes a plurality of characteristic values; Each of the characteristic values ​​is marked as abnormal, and a marking result corresponding to each of the characteristic values ​​is determined, wherein the marking result includes normal and abnormal.

5. The method for identifying abnormal screw tightening according to claim 4, characterized in that: The tightening process includes a screw-in stage and a tightening stage, and the plurality of characteristic values ​​include a maximum torque value of the tightening process, a tightening angle difference value of the tightening process, a climbing angle value of the tightening process, a maximum torque value of the screw-in stage, a torque mean value of the screw-in stage, and a torque distribution kurtosis of the tightening stage; The step of marking each of the characteristic values ​​as abnormal and determining a marking result corresponding to each of the characteristic values ​​includes: Based on the comparison results of the maximum torque value of the tightening process, the tightening angle difference value of the tightening process, and the climbing angle value of the tightening process with the corresponding preset thresholds, determine the marking result corresponding to each of the characteristic values ​​of the maximum torque value of the tightening process, the tightening angle difference value of the tightening process, and the climbing angle value of the tightening process; Based on the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage corresponding to all the sample screws, determine the marking result corresponding to each of the characteristic values ​​in the maximum torque value of the screwing-in stage and the mean torque value of the screwing-in stage by triple standard deviation method; Based on the torque distribution kurtosis of all the tightening stages, a labeling result corresponding to each characteristic value in the torque distribution kurtosis of all the tightening stages is determined by a cluster analysis method.

6. The method for identifying abnormal screw tightening according to claim 2, characterized in that: Before training and verifying the abnormality recognition model based on the feature data of the sample screw, the labeling result corresponding to the feature data, and the category of the sample screw, the method further includes: Based on the feature data of the negative sample screws and the marking results corresponding to the feature data, the negative sample screws are oversampled to obtain feature data of newly added negative sample screws and the marking results corresponding to the feature data, wherein the deviation between the sum of the number of the negative sample screws and the newly added negative sample screws and the number of the positive sample screws meets a preset condition.

7. The method for identifying abnormal screw tightening according to claim 2, characterized in that: Before determining the category of the sample screw based on the tightening data of the sample screw, the method further includes: The tightening data of the sample screws are cleaned to remove abnormal samples from the sample screws.

8. The method for identifying abnormal screw tightening according to claim 2, characterized in that: The characteristic data includes a plurality of characteristic values; the method further includes: If the recognition result indicates that there is an abnormality in the tightening process of the screw to be tested, obtaining decision path information of the feature data in the abnormality recognition model; Based on the decision path information, feature values ​​of the feature data for which a marking result is abnormal are determined.

9. A screw tightening abnormality identification device, characterized in that: The device comprises: A data acquisition module is used to obtain the tightening data of the screw to be tested during the tightening process; A data cleaning module, used to remove abnormal data from the tightening data and determine the data to be tested; A feature extraction module is used to extract features from the data to be tested and determine feature data; The abnormality recognition module is used to input the characteristic data into an abnormality recognition model to obtain the recognition result of the screw to be tested, and the recognition result is used to characterize whether there is an abnormality in the tightening process of the screw to be tested. The abnormality recognition model is obtained based on the characteristic data of the sample screw, the marking result corresponding to the characteristic data of the sample screw and the category training of the sample screw.

10. An electronic device, characterized in that: The electronic device comprises a memory and a processor; The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is performed.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 8 is executed.

Images