Abnormality detection method and device for cell entering shell, electronic equipment and storage medium

Abstract

The embodiment of the application discloses an abnormality detection method and device for cell entering a shell, electronic equipment and a storage medium. The method comprises the following steps: acquiring detection data in the process that a cell entering shell machine drives the cell to enter a battery shell; generating one or more to-be-detected curves according to the detection data; detecting the one or more to-be-detected curves respectively according to historical detection data to obtain a detection result corresponding to each to-be-detected curve; and determining an abnormality result of the cell entering the shell according to the detection result corresponding to each to-be-detected curve. The embodiment of the application can improve the accuracy of the abnormal cell judgment. The above judgment process does not need manual inspection, improves the intelligent degree of the cell detection process, can effectively determine the main abnormal problems in the production process, and can be solved in a targeted manner, thereby improving the accuracy and efficiency of the abnormal detection of the lithium cell entering the shell process.

Description

Technical Field

[0001] This application relates to the field of industrial testing technology, specifically to a method, apparatus, electronic device, and storage medium for detecting abnormalities in battery cell casing. Background Technology

[0002] With the continuous development of industrial products, the requirements for quality inspection of industrial products are also becoming increasingly stringent. In the current battery assembly process, battery cells need to be installed into the battery casing using a casing machine. During this process, abnormalities such as jamming or burrs may occur, leading to a reduction in cell quality and affecting the performance and safety of the finished battery. Currently, the method of monitoring the thrust value during cell installation and manually inspecting cells with abnormal thrust values ​​to determine if an abnormality exists is inaccurate and lacks a high degree of automation. Summary of the Invention

[0003] This application discloses an anomaly detection method, device, electronic device, and storage medium for battery cell installation, which can improve the intelligence and accuracy of anomaly detection during the battery cell installation process.

[0004] The first aspect of this application provides a method for detecting abnormalities in the insertion of battery cells into the casing, the method comprising:

[0005] Acquire detection data during the process of the battery cell being pushed into the battery casing;

[0006] One or more curves to be detected are generated based on the detection data;

[0007] Based on historical testing data, the one or more curves to be tested are tested respectively to obtain the test result corresponding to each curve to be tested; the historical testing data is the test data when no abnormality occurred during the process of the battery cell being pushed into the battery casing in the past.

[0008] The abnormal results of the cell casing are determined based on the detection results corresponding to each of the curves to be detected.

[0009] As an optional implementation, in the first aspect of the present application, the detection data includes the displacement value and / or thrust value of the battery cell being pushed into the battery casing;

[0010] The step of generating one or more curves to be detected based on the detection data includes:

[0011] Generate a detection curve corresponding to the thrust value based on the thrust value, and / or generate a detection curve corresponding to the displacement value based on the displacement value.

[0012] As an optional implementation, in the first aspect of the embodiments of this application, the step of detecting one or more curves to be detected based on historical detection data to obtain a detection result corresponding to each curve to be detected includes:

[0013] A standard curve is generated for each curve to be detected based on the historical detection data, and a slope threshold is determined for each standard curve based on each standard curve.

[0014] The slope of each of the curves to be detected is determined, and the slope of each curve to be detected is compared with the corresponding slope threshold to obtain the slope comparison result of each curve to be detected.

[0015] Determine the similarity between each of the curves to be detected and the corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result of each curve to be detected;

[0016] Based on the slope comparison results and similarity comparison results of each of the curves to be detected, the detection result corresponding to each curve to be detected is obtained.

[0017] As an optional implementation, in the first aspect of the embodiments of this application, the method further includes:

[0018] Based on the detection data, determine the abnormal data corresponding to each of the curves to be detected;

[0019] The step of detecting one or more curves to be detected based on historical detection data to obtain a detection result for each curve to be detected includes:

[0020] Determine the threshold for abnormal data based on historical detection data;

[0021] The abnormal data corresponding to each of the curves to be detected is compared with the abnormal data threshold to obtain the detection result corresponding to each of the curves to be detected.

[0022] As an optional implementation, in the first aspect of the embodiments of this application, the abnormal data includes stuttering values ​​and glitch values, and the abnormal data threshold includes stuttering threshold and glitch threshold;

[0023] The step of comparing the abnormal data corresponding to each of the curves to be detected with an abnormal data threshold to obtain the detection result corresponding to each of the curves to be detected includes:

[0024] The stuttering value corresponding to each of the curves to be detected is compared with the stuttering threshold to obtain the stuttering comparison result corresponding to each of the curves to be detected.

[0025] The burr value corresponding to each of the curves to be detected is compared with the burr threshold to obtain the burr comparison result corresponding to each of the curves to be detected.

[0026] Based on the stuttering comparison result and the glitch comparison result corresponding to each of the curves to be detected, the detection result corresponding to each of the curves to be detected is obtained.

[0027] As an optional implementation, in the first aspect of the embodiments of this application, the method further includes:

[0028] Based on the stuttering comparison results corresponding to the target detection curve, a first number of target detection curves with stuttering values ​​greater than the stuttering threshold are determined; the target detection curve is any of the detection curves.

[0029] Based on the burr comparison results corresponding to the target detection curve, a second number of burr values ​​in the target detection curve that are greater than the burr threshold is determined;

[0030] Based on the first quantity, the weight of the stuttering comparison result, the second quantity, and the weight of the glitch comparison result, the anomaly score corresponding to the target detection curve is determined; wherein, the weight of the stuttering comparison result is greater than the weight of the glitch comparison result.

[0031] The abnormality level of the battery cell casing is determined based on the abnormality score corresponding to each of the curves to be detected.

[0032] As an optional implementation, in a first aspect of the present application, after generating one or more curves to be detected based on the detection data, the method further includes:

[0033] According to the battery cell assembly process, each of the curves to be tested is divided into multiple sub-curves; each sub-curve corresponds one-to-one with a sub-process of the battery cell assembly process.

[0034] The step of detecting one or more curves to be detected based on historical detection data to obtain a detection result for each curve to be detected includes:

[0035] Each sub-curve of each curve to be detected is detected according to historical detection data to obtain the detection result corresponding to each sub-curve of each curve to be detected, and the detection result corresponding to each curve to be detected is determined according to the detection result corresponding to each sub-curve.

[0036] A second aspect of this application provides an abnormal detection device for battery cell insertion, the device comprising:

[0037] The data acquisition module is used to acquire detection data of the battery cell during the process of being pushed into the battery casing.

[0038] A curve construction module is used to generate one or more curves to be detected based on the detection data;

[0039] The curve detection module is used to detect one or more curves to be detected based on historical detection data, and to obtain the detection result corresponding to each curve to be detected; the historical detection data refers to the detection data of the battery cells when no abnormality occurred during the process of the battery cells being pushed into the battery casing.

[0040] An anomaly detection module is used to determine the abnormal result of the cell insertion into the casing based on the detection result corresponding to each of the curves to be detected.

[0041] A third aspect of this application provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor enables the processor to implement any of the abnormal detection methods for battery cell installation disclosed in the embodiments of this application.

[0042] A fourth aspect of this application provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements any of the abnormal detection methods for battery cell installation disclosed in the embodiments of this application.

[0043] Compared with related technologies, the embodiments of this application have the following beneficial effects:

[0044] The process involves acquiring detection data during the process of a battery cell being pushed into the battery casing. Based on this data, one or more test curves are generated. Historical detection data—data from previous instances where no abnormalities occurred during the cell's insertion into the battery casing—is used to test each of the generated test curves, yielding a test result for each curve. Finally, the abnormal cell insertion result is determined based on the test result for each curve. Implementing this embodiment allows for a clear visualization of the cell's insertion process by generating test curves. Furthermore, comparing and analyzing historical detection data with the test curves, and synthesizing the test results of each curve, improves the accuracy of identifying abnormal cells. This judgment process eliminates the need for manual inspection, enhancing the intelligence of the cell testing process. Additionally, analyzing historical detection data with the test curves effectively identifies major anomalies in the production process, enabling targeted solutions and improving the accuracy and efficiency of anomaly detection during the lithium-ion battery cell insertion process. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram illustrating an application scenario of an abnormal detection method for battery cell installation disclosed in one embodiment;

[0047] Figure 2 This is a schematic flowchart of an abnormal detection method for battery cell casing disclosed in one embodiment;

[0048] Figure 3 This is a schematic flowchart of another method for detecting abnormalities in battery cell casing, as disclosed in one embodiment.

[0049] Figure 4 This is a schematic flowchart of another method for detecting abnormalities in battery cell casing, as disclosed in one embodiment.

[0050] Figure 5 This is a schematic flowchart of another method for detecting abnormalities in battery cell casing, as disclosed in one embodiment.

[0051] Figure 6 This is a schematic diagram of a curve anomaly detection process disclosed in one embodiment;

[0052] Figure 7 This is a schematic diagram of a stuttering and glitch detection process disclosed in one embodiment;

[0053] Figure 8 This is a schematic diagram of the structure of an abnormal detection device for battery cell casing disclosed in one embodiment;

[0054] Figure 9 This is an architecture diagram of an abnormality detection system for battery cell casing disclosed in one embodiment;

[0055] Figure 10 This is a schematic diagram of the structure of an electronic device disclosed in one embodiment. Detailed Implementation

[0056] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0058] In existing battery assembly processes, battery cells need to be installed into the battery casing using a casing machine. Current methods for detecting cell anomalies rely on monitoring the thrust value during cell installation and manually inspecting cells with abnormal thrust values ​​to determine if an anomaly exists. This method suffers from poor accuracy and low levels of automation. This application discloses a method, apparatus, electronic device, and storage medium for detecting anomalies during battery cell installation, which can improve the accuracy of anomaly detection. These will be described in detail below.

[0059] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application scenario of an anomaly detection method for battery cell installation disclosed in one embodiment. For example... Figure 1 As shown, the device may include an electronic device 10, a battery cell 20, and a pushing device 30. The electronic device 10 may be equipped with a processor, but is not limited thereto. The pushing device 30 may be equipped with sensing devices such as an encoder and a pressure sensor to collect detection data during the process of the pushing device 30 pushing the battery cell 20 into the battery casing. The electronic device 10 can be communicatively connected to the pushing device 30. Alternatively, the electronic device 10 can be communicatively connected to an external detection device, which is used to collect detection data during the process of the pushing device 30 pushing the battery cell 20 into the battery casing.

[0060] The electronic device 10 can acquire detection data during the process of the battery cell 20 being pushed into the battery casing by the pushing device 30 or an external detection device, and generate one or more test curves based on the detection data. Then, the electronic device 10 performs detection on the generated one or more test curves according to historical detection data, obtaining a detection result corresponding to each test curve; wherein, the historical detection data refers to the detection data when no abnormalities occurred during the process of the battery cell being pushed into the battery casing by the pushing device 30 in the past. Finally, the electronic device 10 determines the abnormal result of the battery cell 20 entering the casing based on the detection result corresponding to each test curve.

[0061] Please see Figure 2 , Figure 2 This is a schematic flowchart of a method for detecting abnormalities in the casing of a battery cell, as disclosed in one embodiment. This method can be applied to the aforementioned electronic device 10. Figure 2As shown, the method may include the following steps:

[0062] 210. Obtain detection data during the process of the battery cell being pushed into the battery casing.

[0063] In this embodiment, the battery cell can be pushed into the battery casing by a pushing device in the casing insertion machine. The pushing device may have a built-in sensing device, such as an encoder and a pressure sensor. The pushing device collects detection data during the process of pushing the battery cell into the battery casing using the built-in sensing device and sends the collected detection data to an electronic device, enabling the electronic device to obtain the detection data of the battery cell being pushed into the battery casing. Alternatively, the pushing device may have an external sensing device that collects detection data during the process of the battery cell being pushed into the battery casing and sends the collected detection data to an electronic device, enabling the electronic device to obtain the detection data of the battery cell being pushed into the battery casing. The detection data can reflect the process of the pushing device pushing the battery cell into the battery casing, such as the displacement data of the battery cell, the velocity data of the battery cell, the thrust data of the pushing device, and the angle data of the pushing device pushing the battery cell, etc.

[0064] In this embodiment of the application, the sensing device can collect and transmit detection data according to preset collection rules. For example, the sensing device can continuously collect detection data, or the sensing device can collect detection data at certain time intervals. The sensing device can send the collected detection data to the electronic device in real time, or it can send the collected detection data to the electronic device at certain time intervals.

[0065] 220. Generate one or more curves to be tested based on the test data.

[0066] In this embodiment, the electronic device generates one or more detection curves based on the acquired detection data. If the detection data includes only one data category, the electronic device can generate one detection curve, or it can generate multiple detection curves according to a certain acquisition time length. If the detection data includes more than one data category, the electronic device can generate multiple detection curves according to the number of data categories. The data categories can represent different types of data, such as displacement data, thrust data, velocity data, and angle data. The detection curve is a curve obtained by fitting line segments to the acquired detection data. The detection curve reflects the changes in the detection data.

[0067] 230. Based on historical test data, one or more curves to be tested are tested respectively to obtain the test results corresponding to each curve to be tested; historical test data are the test data when no abnormality occurred during the process of the battery cell being pushed into the battery casing.

[0068] In this embodiment, after determining that no abnormality occurred during the process of the battery cell being pushed into the battery casing by the pushing device, the electronic device stores the detection data of the process of the pushing device pushing the battery cell into the battery casing into a database as historical detection data. The database can be an internal database of the electronic device or an external database.

[0069] The electronic device uses stored historical detection data to detect one or more curves generated based on the detection data, obtaining a detection result for each curve. Specifically, the electronic device can plot a standard curve based on historical detection data and compare this standard curve with each curve to obtain a detection result. For example, if a curve differs significantly from the standard curve, its detection result is considered abnormal. The electronic device can also compare historical detection data with each peak (or the highest peak) and / or each trough (the lowest trough) of each curve to obtain a detection result. For example, if the highest peak of a curve is greater than that of historical detection data, its detection result is considered abnormal. The standard curve reflects the changes in historical detection data.

[0070] 240. Determine the abnormal results of cell casing based on the test results corresponding to each test curve.

[0071] In this embodiment, the electronic device can determine the abnormal result of the battery cell assembly process based on the detection result corresponding to each detection curve. Specifically, if only one detection curve is generated, and the detection result corresponding to that curve is abnormal, then the battery cell assembly process is determined to be abnormal; if the detection result corresponding to that curve is normal, then the battery cell assembly process is determined to be normal. If more than one detection curve is generated, the electronic device can determine that the battery cell assembly process is abnormal when the detection result corresponding to any one of the detection curves is abnormal; the electronic device can also determine that the battery cell assembly process is abnormal only when the detection results corresponding to all the detection curves are abnormal; or the electronic device can determine that the battery cell assembly process is abnormal only when the detection results corresponding to more than a certain number of detection curves are abnormal.

[0072] Using the above embodiments, the generated test curve can intuitively reflect the specific situation of the cell being pushed into the battery casing. By comparing and analyzing the test curves with historical test data and combining the test results of each test curve, the accuracy of judging abnormal cells can be improved. The above judgment process does not require manual inspection, which improves the intelligence level of the cell testing process. In addition, by analyzing the test curves with historical test data, the main abnormal problems in the production process can be effectively identified for targeted solutions, thereby improving the accuracy and efficiency of abnormal detection in the lithium battery cell casing process.

[0073] In some embodiments, the process by which an electronic device performs detection on one or more curves to be detected based on historical detection data and obtains a detection result for each curve to be detected may include the following steps:

[0074] Generate a standard curve corresponding to each curve to be tested based on historical test data, and determine the slope threshold corresponding to each standard curve based on each standard curve.

[0075] Determine the slope of each curve to be detected, and compare the slope of each curve to be detected with the corresponding slope threshold to obtain the slope comparison result of each curve to be detected.

[0076] Determine the similarity between each curve to be detected and its corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result for each curve to be detected;

[0077] Based on the slope comparison results and similarity comparison results of each curve to be detected, the detection result corresponding to each curve to be detected is obtained.

[0078] In this embodiment, the electronic device can generate one or more standard curves based on historical testing data. Each standard curve corresponds one-to-one with the curve to be tested. For example, if the electronic device generates three curves to be tested based on testing data of three different data categories, it can then generate three standard curves based on historical testing data of the corresponding data categories. Next, the electronic device determines the slope threshold corresponding to each standard curve. Specifically, for each standard curve, the electronic device can generate multiple base curves (e.g., 500) based on several historical testing data points. It calculates the slope of each base curve and sorts them in ascending or descending order. The base curves with slopes distributed in the middle 50% are selected, and the average value of these base curves is calculated as the slope threshold of the standard curve. The standard curve is then fitted using a numerical method. Here, the base curve is a curve generated based on the testing data of any historical battery cell.

[0079] The electronic device determines the slope of each curve to be detected and compares the slope of each curve to be detected with the slope threshold of the corresponding standard curve to obtain the slope comparison result of each curve to be detected. In specific implementation, for a curve to be detected, if the slope of the curve to be detected is less than the slope threshold of the corresponding standard curve, the slope comparison result of the curve to be detected is determined to be abnormal; or, if the slope of the curve to be detected is greater than the slope threshold of the corresponding standard curve, the slope comparison result of the curve to be detected is determined to be abnormal; no specific limitation is made here.

[0080] The electronic device can determine the similarity between each curve to be detected and its corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result for each curve to be detected. In practice, the similarity threshold is an empirical value. The electronic device can determine the similarity between each curve to be detected and its corresponding standard curve using methods such as point-based methods, shape-based methods, or segmentation-based methods, and compare each similarity with a similarity threshold or a corresponding similarity threshold to obtain the similarity comparison result for each curve to be detected. For a curve to be detected, if the similarity of the curve to be detected is less than the corresponding similarity threshold, the slope comparison result of the curve to be detected is determined to be abnormal.

[0081] The electronic device can obtain the slope comparison results and similarity comparison results for each curve to be detected using the above method. Based on these two results, the detection result for each curve to be detected is determined. In specific implementation, for a given curve to be detected, if either the slope comparison result or the similarity comparison result is abnormal, the detection result corresponding to that curve is determined to be abnormal; alternatively, if both the slope comparison result and the similarity comparison result are abnormal, the detection result corresponding to that curve is determined to be abnormal.

[0082] By constructing a standard curve and comparing the slope and overall similarity between the curve to be detected and the corresponding standard curve, the accuracy of the analysis of the curve to be detected can be further improved, and the analysis and detection of the curve to be detected can be performed more efficiently, thereby improving the efficiency of the entire anomaly detection process.

[0083] In one embodiment, see Figure 3 , Figure 3 This is a schematic flowchart of another method for detecting abnormalities in the battery cell casing, as disclosed in one embodiment. This method can be applied to the aforementioned electronic device 10. Figure 3 As shown, the method may include the following steps:

[0084] 310. Obtain the displacement and / or thrust value of the cell being pushed into the battery casing.

[0085] In this embodiment, the electronic device can acquire the displacement value of the battery cell being pushed into the battery casing by the pushing device, and the thrust value of the pushing device pushing the battery cell into the battery casing, through a sensing device such as an encoder, and through a sensing device such as a pressure sensor. The electronic device can acquire the thrust value and displacement value according to a preset acquisition rule, which is the same as the detection data acquisition process described above, and will not be repeated here.

[0086] 320. Generate the curve to be detected corresponding to the thrust value based on the thrust value, and / or generate the curve to be detected corresponding to the displacement value based on the displacement value.

[0087] In this embodiment, the electronic device can generate a detection curve corresponding to the thrust value and a detection curve corresponding to the displacement value. Specifically, if the electronic device acquires a thrust value or a displacement value, it should generate a detection curve corresponding to the thrust value or a detection curve corresponding to the displacement value; if the electronic device acquires both thrust and displacement values, it can generate a detection curve corresponding to the thrust value and / or a detection curve corresponding to the displacement value.

[0088] 330. Based on historical test data, test the curve to be tested corresponding to the thrust value and / or the curve to be tested corresponding to the displacement value to obtain the test result corresponding to each curve to be tested; historical test data are the test data when no abnormality occurred during the process of the battery cell being pushed into the battery casing.

[0089] In this embodiment of the application, the historical detection data corresponding to the detection data may include historical thrust value and historical displacement value. The historical thrust value is the thrust data when no abnormality occurs during the process of the pushing device pushing the historical cell into the battery casing. The historical displacement value is the displacement data when no abnormality occurs during the process of the historical cell being pushed into the battery casing by the pushing device.

[0090] The electronic equipment performs corresponding detections on the curves to be detected corresponding to thrust values ​​and / or displacement values ​​based on historical detection data, thereby obtaining the detection result for each curve to be detected. Specifically, for the curve to be detected corresponding to a thrust value, the electronic equipment can use historical thrust values ​​to detect the curve to obtain the detection result. Specifically, the electronic equipment can plot a historical thrust value curve based on historical thrust values ​​and compare this historical thrust value curve with the curve to be detected corresponding to the thrust value to obtain the detection result. Alternatively, the electronic equipment can compare the historical thrust value with the thrust value of each peak (or the highest peak) and / or the thrust value of each trough (the lowest trough) in the curve to be detected corresponding to the historical thrust value, and / or the number of peaks / troughs in the curve to be detected corresponding to the thrust value to obtain the detection result for the curve to be detected corresponding to the thrust value. Similarly, for the curve to be detected corresponding to the displacement value, the electronic device can use historical displacement values ​​to detect the curve to be detected corresponding to the displacement value, and obtain the detection result of the curve to be detected corresponding to the displacement value; wherein, the electronic device can draw a historical displacement value curve based on the historical displacement values, and compare the historical displacement value curve with the curve to be detected corresponding to the displacement value to obtain the detection result of the curve to be detected corresponding to the displacement value; or, the electronic device can compare the historical displacement value with the displacement value of each peak (or the highest peak) in the curve to be detected corresponding to the historical displacement value, and / or the displacement value of each trough (the lowest trough), and / or the number of peaks / troughs in the curve to be detected corresponding to the displacement value, to obtain the detection result of the curve to be detected corresponding to the displacement value.

[0091] 340. Determine the abnormal results of cell casing based on the test results corresponding to each test curve.

[0092] In this embodiment, if only the test curve corresponding to the thrust value is generated, and the test result of the test curve corresponding to the thrust value is abnormal, then the cell insertion process is determined to be abnormal; if the test result of the test curve corresponding to the thrust value is normal, then the cell insertion process is determined to be normal. If only the test curve corresponding to the displacement value is generated, and the test result of the test curve corresponding to the displacement value is abnormal, then the cell insertion process is determined to be abnormal; if the test result of the test curve corresponding to the displacement value is normal, then the cell insertion process is determined to be normal. If both the test curve corresponding to the thrust value and the test curve corresponding to the displacement value are generated, then the electronic device can determine that the cell insertion process is abnormal if either the test result of the test curve corresponding to the thrust value is abnormal or the test result of the test curve corresponding to the displacement value is abnormal; alternatively, the electronic device can determine that the cell insertion process is abnormal only if both the test results of the test curve corresponding to the thrust value and the test results of the test curve corresponding to the displacement value are abnormal.

[0093] Using the above embodiments, pressure and / or displacement values ​​are used as detection data, which can better reflect the process conditions during the process of pushing the battery cell into the battery casing. By generating corresponding detection curves based on the pressure and / or displacement values, and detecting these curves according to historical detection data, the abnormal results of the battery cell insertion can be determined more accurately, thus improving the accuracy of abnormal cell insertion detection.

[0094] In some embodiments, the process by which the electronic device performs detection on the curve to be detected corresponding to the thrust value and / or the curve to be detected corresponding to the displacement value based on historical detection data, and obtains the detection result for each curve to be detected, may include the following steps:

[0095] Generate standard curves corresponding to thrust values ​​and / or displacement values ​​based on historical test data, and determine the slope threshold for each standard curve based on each standard curve.

[0096] Determine the slope of each curve to be detected, and compare the slope of each curve to be detected with the corresponding slope threshold to obtain the slope comparison results of the curve to be detected corresponding to the thrust value and / or the slope comparison results of the curve to be detected corresponding to the displacement value.

[0097] Determine the similarity between each curve to be detected and its corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison results of the curve to be detected corresponding to the thrust value and / or the similarity comparison results of the curve to be detected corresponding to the displacement value.

[0098] Based on the slope comparison results and similarity comparison results of each curve to be detected, the detection result corresponding to each curve to be detected is obtained.

[0099] In this embodiment, the historical detection data corresponding to the detection data may include historical thrust values ​​and historical displacement values. The electronic device can generate a standard curve corresponding to the historical thrust values, and / or generate a standard curve corresponding to the historical displacement values. Then, the electronic device determines the slope threshold corresponding to each standard curve.

[0100] The electronic device determines the slope of the curve to be tested corresponding to the thrust value, and compares the slope of the curve to be tested corresponding to the thrust value with the slope threshold of the standard curve corresponding to the thrust value to obtain the slope comparison result of the curve to be tested corresponding to the thrust value. And / or, the electronic device determines the slope of the curve to be tested corresponding to the displacement value, and compares the slope of the curve to be tested corresponding to the displacement value with the slope threshold of the standard curve corresponding to the displacement value to obtain the slope comparison result of the curve to be tested corresponding to the displacement value.

[0101] The electronic device can determine the similarity between the curve to be detected corresponding to the thrust value and the standard curve corresponding to the thrust value, and compare the similarity with a similarity threshold to obtain the similarity comparison result of the curve to be detected corresponding to the thrust value. And / or, the electronic device can determine the similarity between the curve to be detected corresponding to the displacement value and the standard curve corresponding to the displacement value, and compare the similarity with a similarity threshold to obtain the similarity comparison result of the curve to be detected corresponding to the displacement value.

[0102] The electronic device can obtain the slope comparison results and similarity comparison results of the curve to be detected corresponding to the thrust value, and / or the slope comparison results and similarity comparison results of the curve to be detected corresponding to the displacement value, through the above methods. The electronic device can jointly determine the detection result of the curve to be detected corresponding to the thrust value based on the slope comparison results and similarity comparison results. Similarly, the electronic device can jointly determine the detection result of the curve to be detected corresponding to the displacement value based on the slope comparison results and similarity comparison results.

[0103] In practice, for a given curve to be detected, if the slope comparison result or the similarity comparison result is abnormal, the detection result corresponding to the curve to be detected is determined to be abnormal; or, the detection result corresponding to the curve to be detected is determined to be abnormal only when both the slope comparison result and the similarity comparison result are abnormal.

[0104] By constructing a standard curve and / or displacement value corresponding to the thrust value and comparing the slope and overall similarity between the curve to be detected and the corresponding standard curve, the accuracy of the analysis of the curve to be detected can be further improved, and the curve to be detected can be analyzed and detected more efficiently, thereby improving the efficiency of the entire cell casing anomaly detection process.

[0105] In one embodiment, see Figure 4 , Figure 4 This is a schematic flowchart of another method for detecting abnormalities in the battery cell casing, as disclosed in one embodiment. This method can be applied to the aforementioned electronic device 10. Figure 4 As shown, the method may include the following steps:

[0106] 410. Obtain detection data during the process of the battery cell being pushed into the battery casing.

[0107] 420. Generate one or more curves to be tested based on the test data.

[0108] 430. Based on the detection data, identify the abnormal data corresponding to each curve to be detected.

[0109] In this embodiment, the electronic device determines the abnormal data corresponding to each test curve based on the detection data used to generate the test curve. Specifically, the electronic device can match the data characteristics of the detection data with different data characteristics corresponding to the abnormal data. Each data characteristic corresponding to the abnormal data can correspond to one abnormal data point. Each test curve can correspond to one or more abnormal data points.

[0110] 440. Determine the threshold for abnormal data based on historical detection data.

[0111] In this embodiment, the abnormal data threshold can be determined based on historical detection data. For example, the electronic device can match the data characteristics of historical detection data with different data characteristics corresponding to abnormal data, and determine the abnormal data corresponding to the data characteristic with the highest matching degree as the abnormal data threshold. Alternatively, historical detection data that meets the data characteristics of abnormal data can be filtered from the historical detection data, and the abnormal data threshold can be determined from the filtered historical detection data by selecting the maximum value or calculating the average value.

[0112] 450. Compare the abnormal data corresponding to each curve to be detected with the abnormal data threshold to obtain the detection result corresponding to each curve to be detected.

[0113] In this embodiment, the electronic device compares the abnormal data corresponding to each curve to be detected with an abnormal data threshold to obtain the detection result for each curve. Specifically, if the abnormal data corresponding to a curve to be detected is greater than the abnormal data threshold, the electronic device can determine the detection result for that curve to be detected as abnormal; or, if the abnormal data corresponding to a curve to be detected is less than the abnormal data threshold, the electronic device can determine the detection result for that curve to be detected as abnormal; no specific limitations are made here.

[0114] Using the above embodiments, abnormal data corresponding to the curve to be detected is determined, and the detection result corresponding to each curve to be detected is determined based on the comparison result between the abnormal data corresponding to the curve to be detected and the abnormal data threshold determined by historical detection data. By comparing specific data, the abnormal situation of cell insertion can be judged more accurately, thereby improving the accuracy of cell insertion anomaly judgment.

[0115] In some embodiments, abnormal data includes stuttering values ​​and glitches, and abnormal data thresholds include stuttering thresholds and glitches thresholds.

[0116] The step of the electronic device comparing the abnormal data corresponding to each curve to be detected with an abnormal data threshold to obtain the detection result for each curve to be detected may include the following steps:

[0117] The stuttering value corresponding to each curve to be detected is compared with the stuttering threshold to obtain the stuttering comparison result for each curve to be detected.

[0118] The spur value corresponding to each curve to be detected is compared with the spur threshold to obtain the spur comparison result for each curve to be detected.

[0119] Based on the stuttering comparison results and the glitch comparison results corresponding to each curve to be tested, the detection results corresponding to each curve to be tested are obtained.

[0120] In this embodiment, the electronic device determines the stuttering and burr values ​​of the curve to be tested based on the detection data, such as thrust and displacement values, used to generate the curve. Specifically, the electronic device can detect the peaks of the curve to be tested and match the peaks with the curve features of burrs and stutters. If the match is successful, the stuttering and burr values ​​of the curve to be tested are determined based on the detection data corresponding to the peak. Each curve to be tested may include several stuttering and burr values.

[0121] The electronic device determines the glitch threshold and the stutter threshold based on historical test data. Specifically, the electronic device can generate a standard curve based on the historical test data, and determine the stutter and glitch values ​​of the standard curve using the method described above for determining the stutter and glitch values ​​of the curve to be tested. For each stutter value, a stutter threshold is determined, and for each glitch value, a glitch threshold is determined. For example, the maximum stutter value of the standard curve is determined as the stutter threshold, and the maximum glitch value of the standard curve is determined as the glitch threshold.

[0122] The electronic device compares the stutter value corresponding to each curve to be detected with a stutter threshold to obtain a stutter comparison result for each curve; and compares the glitch value corresponding to each curve to be detected with a glitch threshold to obtain a glitch comparison result for each curve. Based on the stutter comparison result and the glitch comparison result for each curve, the electronic device obtains the detection result for each curve.

[0123] In practical implementation, for a given curve to be tested, if the curve has only one stutter value, the electronic device can determine the stutter comparison result based on that stutter value. If the curve contains more than one stutter value, the electronic device can determine the stutter comparison result based on the number of stutter values ​​that do not meet the stutter value threshold. For example, if more than half of the stutter values ​​are greater than the stutter value threshold, the stutter comparison result is abnormal. Similarly, for a given curve to be tested, if the curve has only one glitch value, the electronic device can determine the glitch comparison result based on that glitch value. If the curve contains more than one glitch value, the electronic device can determine the glitch comparison result based on the number of glitch values ​​that do not meet the glitch value threshold.

[0124] The electronic device obtains the detection result for each curve based on the stuttering comparison result and the glitch comparison result corresponding to each curve to be detected. In specific implementation, for a curve to be detected, if the stuttering comparison result or the glitch comparison result corresponding to the curve to be detected is abnormal, then the detection result corresponding to the curve to be detected is determined to be abnormal; or, if both the stuttering comparison result and the glitch comparison result corresponding to the curve to be detected are abnormal, then the detection result corresponding to the curve to be detected is determined to be abnormal.

[0125] Since burrs and jamming are the main anomalies in the battery cell assembly process, the above embodiment is adopted to determine the detection result corresponding to each curve to be tested by comparing the burr value and the jamming value. This can more accurately determine the abnormal state of the battery cell casing process and improve the accuracy of the battery cell casing anomaly judgment.

[0126] 460. Determine the abnormal results of cell casing based on the test results corresponding to each test curve.

[0127] In some embodiments, the anomaly detection method may further include the following steps:

[0128] Based on the stuttering comparison results corresponding to the target detection curve, determine the first number of target detection curves with stuttering values ​​greater than the stuttering threshold; the target detection curve can be any detection curve.

[0129] Based on the comparison results of the burrs corresponding to the target curve to be detected, determine the second number of burr values ​​in the target curve to be detected that are greater than the burr threshold.

[0130] The anomaly score corresponding to the target curve to be detected is determined based on the weight of the first quantity, the weight of the lag comparison result, the weight of the second quantity, and the weight of the spur comparison result; among them, the weight of the lag comparison result is greater than the weight of the spur comparison result.

[0131] The abnormality level of the battery cell casing is determined based on the abnormality score corresponding to each of the curves to be detected.

[0132] In this embodiment of the application, the number of stuttering values ​​contained in the curve to be detected can be one or more, and the number of glitch values ​​contained can also be one or more. For any curve to be detected, that is, the target curve to be detected, after the electronic device determines the corresponding stuttering comparison result based on the stuttering values ​​contained in the target curve to be detected, it can count the number of stuttering values ​​in the target curve to be detected that are greater than the stuttering threshold, which is the first number; similarly, for the target curve to be detected, after the electronic device determines the corresponding glitch comparison result based on the glitch values ​​contained in the curve to be detected, it can count the number of glitch values ​​in the target curve to be detected that are greater than the glitch threshold, which is the second number.

[0133] Next, the electronic device can calculate the anomaly score corresponding to the target curve to be detected based on the weights of the first quantity, the second quantity, the stuttering comparison result, and the burr comparison result, and determine the anomaly level of the cell casing based on the anomaly score corresponding to each curve to be detected. Among these, since stuttering is a more abnormal situation in the cell casing process, the weight of the stuttering comparison result is greater than the weight of the burr comparison result.

[0134] In practice, the electronic device counts the first and second counts of the curve to be tested, a, as 1. The weight of the stuttering comparison result is 2, and the weight of the glitch comparison result is 1. Therefore, the anomaly score corresponding to the curve to be tested, a, is 1*2 + 1*1 = 3. The electronic device counts the first and second counts of the curve to be tested, b, as 2. Therefore, the anomaly score corresponding to the curve to be tested, b, is 2*2 + 2*1 = 7. Different anomaly levels correspond to different anomaly score ranges. The electronic device can determine the anomaly level of the battery cell casing based on the range where the anomaly score of 10 is located.

[0135] By employing the above embodiments, and assigning different weights to the stuttering comparison results and the burr comparison results, the abnormality level of the battery cell assembly can be determined based on the number of stuttering comparison results and the number of burr comparison results during the battery cell assembly process. This allows for a more effective analysis of the abnormality level of each battery cell assembly, thereby enabling better improvement of existing process issues and further enhancing the intelligence level of the battery cell assembly abnormality detection process.

[0136] In one embodiment, see Figure 5 , Figure 5 This is a schematic flowchart of another method for detecting abnormalities in the battery cell casing, as disclosed in one embodiment. This method can be applied to the aforementioned electronic device 10. Figure 5 As shown, the method may include the following steps:

[0137] 510. Obtain detection data during the process of the battery cell being pushed into the battery casing.

[0138] 520. Generate one or more curves to be tested based on the test data.

[0139] 530. According to the battery cell casing process, each test curve is divided into multiple sub-curves; the multiple sub-curves correspond one-to-one with the multiple casing processes in the battery cell casing process.

[0140] 540. Based on historical testing data, each sub-curve of each curve to be tested is tested separately to obtain the testing result corresponding to each sub-curve of each curve to be tested, and the testing result corresponding to each curve to be tested is determined based on the testing result corresponding to each sub-curve. The historical testing data refers to the testing data from previous years when no abnormalities occurred during the process of the battery cell being pushed into the battery casing.

[0141] In this embodiment, the electronic device divides each generated test curve into multiple sub-curves according to the cell casing process flow. Each sub-curve corresponds one-to-one with a sub-process of the cell casing process flow. Specifically, the casing sub-process of the cell casing process flow includes one or more of the following: initial position, breathing column position, casing preparation position, casing mold opening position, and casing completion position. If the multiple casing sub-processes of the cell casing process flow include all of the above-mentioned casing processes, then the electronic device can divide each test curve into 6 sub-curves.

[0142] For each sub-curve, the electronic device can detect the sub-curve based on stored historical detection data to obtain the detection result for each sub-curve. Then, based on the detection results for each sub-curve, the electronic device determines the detection result for each curve to be detected. In specific implementation, for a curve to be detected, it is divided into 6 sub-curves. The electronic device can draw six standard sub-curves based on historical detection data, compare each standard sub-curve with its corresponding sub-curve to obtain the detection result for each sub-curve, and then determine the detection result for the curve to be detected based on the detection results for each sub-curve. For example, if the detection result for one sub-curve of the curve to be detected is abnormal, the electronic device will classify the detection result for that curve as abnormal; or, if the detection results for a certain number of sub-curves of the curve to be detected are abnormal, the electronic device will classify the detection result for that curve as abnormal; or, only when the detection results for all sub-curves of the curve to be detected are abnormal, will the electronic device classify the detection result for that curve as abnormal.

[0143] By using the above embodiments, the anomaly analysis can be performed by dividing the curve to be detected into multiple sub-curves, which can effectively reflect the abnormal process in the shell-entry process and identify the main abnormal problems. Furthermore, the anomaly analysis based on the results of the multi-segment curve anomaly analysis further improves the accuracy of the anomaly detection process.

[0144] 550. Determine the abnormal results of cell insertion based on the test results corresponding to each test curve.

[0145] Please see Figure 6 , Figure 6 This is a schematic diagram of a curve anomaly detection process disclosed in one embodiment. In some embodiments, the electronic device performs the following steps in the process of detecting each sub-curve of each curve to be detected based on historical detection data, obtaining the detection result corresponding to each sub-curve of each curve to be detected, and determining the detection result corresponding to each curve to be detected based on the detection result corresponding to each sub-curve:

[0146] A standard sub-curve is generated for each sub-curve based on historical detection data, and a slope threshold is determined for each standard sub-curve.

[0147] Determine the slope of each sub-curve and compare the slope of each sub-curve with a slope threshold to obtain the slope comparison result of each sub-curve;

[0148] Determine the similarity between each sub-curve and its corresponding standard sub-curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result for each sub-curve;

[0149] Based on the slope comparison results and similarity comparison results of each sub-curve, the detection result corresponding to each curve to be detected is obtained.

[0150] In the embodiments of this application, such as Figure 6 As shown, the electronic device can generate a standard curve based on historical test data, and divide the standard curve into multiple standard sub-curves according to the multiple sub-processes of the cell casing process. Each standard sub-curve corresponds one-to-one with a sub-curve in the curve to be tested.

[0151] The electronic device can determine the maximum slope of each standard sub-curve and, based on each maximum slope, determine a slope threshold using statistical methods such as selecting the maximum, minimum, or average value. For each sub-curve, the electronic device can use the maximum or average slope of the sub-curve as its slope and compare this slope with the slope threshold to obtain the slope comparison result for each sub-curve. In practice, the electronic device can classify the slope comparison results of sub-curves with slopes greater than the slope threshold as abnormal.

[0152] The electronic device calculates the similarity between each sub-curve and its corresponding sub-curve, and compares the obtained similarity with a similarity threshold to obtain the similarity comparison result for each sub-curve. The similarity threshold can be an empirical value. In practice, the electronic device can classify the similarity comparison results of sub-curves with similarity scores below the similarity threshold as abnormal.

[0153] The electronic device can obtain the slope comparison results and similarity comparison results of each sub-curve through the above method, and determine the detection result corresponding to the curve to be detected based on the slope comparison results and similarity comparison results of each sub-curve. In specific implementation, for a curve to be detected, it is divided into 6 sub-curves; if the slope comparison result or similarity comparison result of a sub-curve is abnormal, the detection result corresponding to that sub-curve is determined to be abnormal; or, if both the slope comparison result and the similarity comparison result of a sub-curve are abnormal, then the detection result corresponding to that sub-curve is determined to be abnormal. If the detection result corresponding to one sub-curve of the curve to be detected is abnormal, then the electronic device judges the detection result corresponding to that curve to be detected as abnormal; or, if the detection results corresponding to a certain number (e.g., more than 3) of the sub-curves of the curve to be detected are abnormal, then the electronic device judges the detection result corresponding to that curve to be detected as abnormal; or, if the detection results corresponding to all sub-curves of the curve to be detected are abnormal, then the electronic device judges the detection result corresponding to that curve to be detected as abnormal.

[0154] By constructing a standard curve and comparing the slope and overall similarity between the curve to be detected and the corresponding standard curve, the accuracy of the analysis of the curve to be detected can be further improved, and the analysis and detection of the curve to be detected can be performed more efficiently, thus improving the efficiency of the entire anomaly detection process. Furthermore, anomaly analysis based on the anomaly analysis results of multiple sub-curves also improves the accuracy of the anomaly detection process.

[0155] Please see Figure 7 , Figure 7 This is a schematic diagram of a stuttering and glitch detection process disclosed in one embodiment. In some embodiments, the step of the electronic device performing the following steps—detecting each sub-curve of each target curve according to historical detection data, obtaining the detection result corresponding to each sub-curve of each target curve, and determining the detection result corresponding to each target curve based on the detection result corresponding to each sub-curve—may include the following steps:

[0156] Based on the test data, determine the stuttering and puncture values ​​corresponding to each sub-curve;

[0157] Determine the threshold for abnormal data based on historical detection data;

[0158] The stuttering value corresponding to each sub-curve is compared with the stuttering threshold to obtain the stuttering comparison result for each sub-curve.

[0159] The spurt value corresponding to each sub-curve is compared with the spurt threshold to obtain the spurt comparison result for each sub-curve.

[0160] Based on the stuttering comparison results and the glitch comparison results corresponding to each sub-curve, the detection results corresponding to each curve to be detected are obtained.

[0161] In the embodiments of this application, such as Figure 7 As shown, the electronic device generates the detection data, such as thrust and displacement values, for each sub-curve segment to determine the stuttering and burr values. In practice, the electronic device can detect the peaks of the sub-curves and match them with the curve features of burrs and stutters. If a match is successful, the stuttering and burr values ​​of the sub-curve are determined based on the detection data corresponding to the peak. Each sub-curve can include several stuttering and burr values.

[0162] The electronic device determines the glitch threshold and the stutter threshold based on historical test data. Specifically, the electronic device can generate a standard curve based on the historical test data, and divide the standard curve into multiple standard sub-curves. Following the method described above for determining the stutter and glitch values ​​of the sub-curves, the stutter and glitch values ​​of each standard sub-curve are determined. A stutter threshold is determined based on the stutter value of each standard sub-curve, and a glitch threshold is determined based on the glitch value of each standard sub-curve. For example, the maximum stutter value in each standard sub-curve is determined as the stutter threshold, and the maximum glitch value in each standard sub-curve is determined as the glitch threshold.

[0163] The electronic device compares the stutter value corresponding to each sub-curve with a stutter threshold to obtain a stutter comparison result for each sub-curve; and compares the glitch value corresponding to each sub-curve with a glitch threshold to obtain a glitch comparison result for each sub-curve. Based on the stutter comparison result and the glitch comparison result for each sub-curve, the electronic device determines the detection result corresponding to the curve to be detected.

[0164] In practice, a given curve to be tested is divided into six sub-curves. Each sub-curve can have one or more stuttering values. The electronic device can then determine the stuttering comparison result of the sub-curve based on each stuttering value. In practice, if a stuttering value in a sub-curve does not meet the stuttering threshold, the stuttering comparison result is considered abnormal, and the stuttering comparison result of that sub-curve is also considered abnormal.

[0165] Similarly, each sub-curve can have one or more glitch values, and the electronic device can determine the glitch comparison result of the sub-curve based on each glitch value. In specific implementation, if a glitch value in a sub-curve does not meet the glitch threshold, the glitch comparison result is abnormal, and the glitch comparison result of that sub-curve is also abnormal.

[0166] The electronic device obtains the detection result for each curve to be detected based on the stuttering comparison result and the glitch comparison result corresponding to each sub-curve. Specifically, for a curve to be detected, it is divided into 6 sub-curves. If the stuttering comparison result and / or glitch comparison result of at least one sub-curve is abnormal, then the detection result for that curve to be detected is also abnormal; if both the stuttering and glitch comparison results of each sub-curve are normal, then the detection result for that curve to be detected is also normal. More specifically, if the stuttering comparison result of at least one sub-curve is abnormal and the glitch comparison result of each sub-curve is normal, then the detection result for that curve to be detected is stuttering; if the glitch comparison result of at least one sub-curve is abnormal and the stuttering comparison result of each sub-curve is normal, then the detection result for that curve to be detected is glitch.

[0167] By using the above embodiments, the abnormal conditions of the battery cell are judged by comparing the stuttering value and the burr value, which can more accurately determine the abnormal state of the battery cell in the casing process and improve the accuracy of the abnormal judgment of the battery cell in the casing. Furthermore, the abnormal analysis based on the abnormal analysis results of multi-segment curves also improves the accuracy of the abnormal detection process.

[0168] In some embodiments, the anomaly detection method may further include the following steps:

[0169] Based on the stuttering comparison results corresponding to the target sub-curve, determine the number of stuttering values ​​in the target sub-curve that are greater than the stuttering threshold; the target sub-curve can be any sub-curve.

[0170] Based on the spur comparison results corresponding to the target sub-curve, determine the number of spur values ​​in the target sub-curve that are greater than the spur threshold.

[0171] Based on the number of stutter values ​​greater than the stutter threshold in the target sub-curve, the weight of the stutter comparison result, the number of glitch values ​​greater than the glitch threshold in the target sub-curve, and the weight of the glitch comparison result, the anomaly score corresponding to the target sub-curve is determined, and the anomaly score corresponding to each curve to be detected is determined based on the anomaly score corresponding to each sub-curve; among which, the weight of the stutter comparison result is greater than the weight of the glitch comparison result.

[0172] The abnormality level of the battery cell entering the casing is determined based on the abnormal score corresponding to each curve to be detected.

[0173] In this embodiment, the curve to be detected can be divided into multiple sub-curves. Each sub-curve may contain one or more stuttering values ​​and one or more glitch values. For any sub-curve, i.e., the target sub-curve, after determining the corresponding stuttering comparison result based on the stuttering values ​​contained in the target sub-curve, the electronic device can count the number of stuttering values ​​in the target sub-curve that are greater than the stuttering threshold. Similarly, after determining the corresponding glitch comparison result based on the glitch values ​​contained in the target sub-curve, the electronic device can count the number of glitch values ​​in the target sub-curve that are greater than the glitch threshold.

[0174] Next, the electronic device can calculate the anomaly score corresponding to the target sub-curve based on the number of stutter values ​​exceeding the stutter threshold in the target sub-curve, the weight of the stutter comparison result, the number of glitch values ​​exceeding the glitch threshold in the target sub-curve, and the weight of the glitch comparison result. Based on this anomaly score, the anomaly level of the cell assembly is determined. Since stuttering is a more abnormal situation in the cell assembly process, the weight of the stutter comparison result is greater than the weight of the glitch comparison result.

[0175] In practice, the curve to be tested is divided into 6 sub-curves. The electronic device counts all sub-curves and finds 2 instances where the stutter value exceeds the stutter threshold and 2 instances where the glitch value exceeds the glitch threshold. The weight of the stutter comparison result is 2, and the weight of the glitch comparison result is 1. Therefore, the anomaly score is 2*2 + 2*1 = 7. Different anomaly levels correspond to different anomaly score ranges. The electronic device can determine the anomaly level of the battery cell casing based on the range where the anomaly score of 7 falls.

[0176] By employing the above embodiments, and assigning different weights to the stuttering comparison results and the burr comparison results, the abnormalities in the cell assembly process can be determined based on the number of stuttering comparison results and the number of burr comparison results during the cell assembly process. This allows for a more effective analysis of the degree of abnormality in each cell assembly process, thereby enabling better improvement of existing process issues and further enhancing the intelligence level of the abnormality detection process during cell assembly. Furthermore, the anomaly analysis based on multi-segment curves also improves the accuracy of the anomaly detection process.

[0177] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of an anomaly detection device for battery cell casing disclosed in one embodiment. This anomaly detection device can be applied to electronic device 10. For example... Figure 8 As shown, the anomaly detection device 800 may include: a data acquisition module 810, a curve construction module 820, a curve detection module 830, and an anomaly detection module 840.

[0178] The data acquisition module 810 is used to acquire detection data during the process of the battery cell being pushed into the battery casing;

[0179] Curve construction module 820 is used to generate one or more curves to be detected based on the detection data;

[0180] The curve detection module 830 is used to detect one or more curves to be detected based on historical detection data, and obtain the detection result corresponding to each curve to be detected; the historical detection data is the detection data when no abnormality occurred during the process of the battery cell being pushed into the battery casing in the past.

[0181] The anomaly detection module 840 is used to determine the abnormal results of the battery cell entering the casing based on the detection results corresponding to each curve to be detected.

[0182] In some embodiments, the detection data includes the displacement value and / or thrust value of the cell being pushed into the battery casing.

[0183] Curve building module 820 is also used for:

[0184] Generate the detection curve corresponding to the thrust value based on the thrust value, and / or generate the detection curve corresponding to the displacement value based on the displacement value.

[0185] In some embodiments, the curve detection module 830 is further configured to:

[0186] Generate a standard curve corresponding to each curve to be tested based on historical test data, and determine the slope threshold corresponding to each standard curve based on each standard curve.

[0187] Determine the slope of each curve to be detected, and compare the slope of each curve to be detected with the corresponding slope threshold to obtain the slope comparison result of each curve to be detected.

[0188] Determine the similarity between each curve to be detected and its corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result for each curve to be detected;

[0189] Based on the slope comparison results and similarity comparison results of each curve to be detected, the detection result corresponding to each curve to be detected is obtained.

[0190] In some embodiments, Figure 8 The anomaly detection device shown also includes:

[0191] The data detection module 850 is used to determine the abnormal data corresponding to each curve to be detected based on the detection data;

[0192] The curve detection module 830 is also used for:

[0193] Determine the threshold for abnormal data based on historical detection data;

[0194] The abnormal data corresponding to each curve to be detected is compared with the abnormal data threshold to obtain the detection result for each curve to be detected.

[0195] In some embodiments, abnormal data includes stuttering values ​​and glitches, and abnormal data thresholds include stuttering thresholds and glitches thresholds.

[0196] The curve detection module 830 is also used for:

[0197] The stuttering value corresponding to each curve to be detected is compared with the stuttering threshold to obtain the stuttering comparison result for each curve to be detected.

[0198] The spur value corresponding to each curve to be detected is compared with the spur threshold to obtain the spur comparison result for each curve to be detected.

[0199] Based on the stuttering comparison results and the glitch comparison results corresponding to each curve to be tested, the detection results corresponding to each curve to be tested are obtained.

[0200] In some embodiments, the anomaly detection module 840 is further configured to:

[0201] Based on the stuttering comparison results corresponding to the target detection curve, determine the first number of target detection curves with stuttering values ​​greater than the stuttering threshold; the target detection curve can be any detection curve.

[0202] Based on the comparison results of the burrs corresponding to the target curve to be detected, determine the second number of burr values ​​in the target curve to be detected that are greater than the burr threshold.

[0203] The anomaly score corresponding to the target curve to be detected is determined based on the weight of the first quantity, the weight of the lag comparison result, the weight of the second quantity, and the weight of the spur comparison result; among them, the weight of the lag comparison result is greater than the weight of the spur comparison result.

[0204] The abnormality level of the battery cell entering the casing is determined based on the abnormal score corresponding to each curve to be detected.

[0205] In some embodiments, the curve construction module 820 is further configured to:

[0206] After generating one or more test curves based on the test data, each test curve is divided into multiple sub-curves according to the cell casing process; the multiple sub-curves correspond one-to-one with the multiple casing sub-processes in the cell casing process.

[0207] The curve detection module 830 is also used for:

[0208] Based on historical detection data, each sub-curve of each curve to be detected is detected separately to obtain the detection result corresponding to each sub-curve of each curve to be detected, and the detection result corresponding to each curve to be detected is determined based on the detection result corresponding to each sub-curve.

[0209] Please see Figure 9 , Figure 9 This is an architecture diagram of an abnormality detection system for battery cell installation disclosed in one embodiment. Figure 9 As shown, the system includes a data acquisition layer, a data processing layer, and a data display layer.

[0210] The data acquisition layer obtains displacement data of the battery cell being pushed into the battery casing through an encoder, and thrust data of the battery cell being pushed into the battery casing through a pressure sensor. The data acquisition layer also includes a data acquisition platform, which can collect thrust and displacement data according to preset acquisition rules and variable settings.

[0211] The data acquisition layer transmits displacement and pressure data to the data processing layer via data transmission. The data processing layer includes a detection module, a historical data statistics engine, and a database. The data processing layer first determines if there are any anomalies in the acquisition process, specifically whether devices such as encoders and pressure sensors are malfunctioning. If no anomalies are found, the data is transmitted to the detection module. The detection module first generates the curve to be detected and, based on the detection data used to generate the curve, such as thrust and displacement values, determines the stuttering and burr values ​​of the curve. Then, the stuttering detection engine in the detection module compares the stuttering value corresponding to each curve to a stuttering threshold, obtaining the stuttering comparison result for each curve; similarly, the burr detection engine compares the burr value corresponding to each curve to a burr threshold, obtaining the burr comparison result for each curve.

[0212] Meanwhile, the historical data statistics engine can retrieve historical testing data from the database and generate one or more standard curves based on this data. Each standard curve corresponds one-to-one with the curve to be tested. The data processing layer determines the slope of each curve to be tested and compares its slope with the slope threshold of the corresponding standard curve to obtain the slope comparison result for each curve. Furthermore, the data processing layer determines the similarity between each curve to be tested and its corresponding standard curve and compares each similarity with a similarity threshold to obtain the similarity comparison result for each curve. The data processing layer can also detect the maximum value trend of abnormal pressure data to determine changes during the cell assembly process and whether the cell assembly process is abnormal.

[0213] Based on the above-obtained stuttering comparison results, glitch comparison results, similarity detection results, slope detection results, and maximum value trend detection results, the data processing layer can determine the detection result corresponding to the curve to be detected.

[0214] The data display layer may include multiple terminal devices or display devices to display data trends, test results, and parameter settings during the shell insertion process and anomaly detection process, so as to realize real-time monitoring of anomaly detection results and ensure the efficiency of the shell insertion process.

[0215] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of an electronic device disclosed in one embodiment. For example... Figure 10 As shown, the electronic device 1000 may include:

[0216] Memory 1010 storing executable program code;

[0217] Processor 1020 coupled to memory 1010;

[0218] The processor 1020 calls the executable program code stored in the memory 1010 to execute any of the anomaly detection methods disclosed in the embodiments of this application.

[0219] This application discloses a computer-readable storage medium storing a computer program that causes a computer to execute any of the anomaly detection methods disclosed in this application.

[0220] This application discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute any of the anomaly detection methods disclosed in this application.

[0221] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Those skilled in the art should also recognize that the embodiments described in the specification are optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0222] In the various embodiments of this application, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and inherent strategy, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0223] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0224] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0225] If the aforementioned integrated units are implemented as software functional units and sold or used as independent products, they can be stored in a computer-accessible memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in each embodiment of this application.

[0226] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0227] The foregoing has provided a detailed description of an anomaly detection method, apparatus, electronic device, and storage medium disclosed in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for detecting abnormalities in battery cell casing, characterized in that, The method includes: Acquire detection data during the process of the battery cell being pushed into the battery casing; One or more curves to be detected are generated based on the detection data; Based on the detection data, determine the abnormal data corresponding to each of the curves to be detected; The one or more curves to be tested are tested according to historical test data to obtain the test result corresponding to each curve to be tested; the historical test data is the test data when no abnormality occurred during the process of the battery cell being pushed into the battery casing in the past. The abnormal results of the cell casing are determined based on the detection results corresponding to each of the curves to be detected; The abnormal data includes stuttering values ​​and glitch values. The step of detecting one or more curves to be detected based on historical detection data to obtain a detection result for each curve to be detected includes: Anomaly data thresholds are determined based on historical detection data, including stuttering thresholds and glitch thresholds. The stuttering value corresponding to each of the curves to be detected is compared with the stuttering threshold to obtain the stuttering comparison result corresponding to each of the curves to be detected. The burr value corresponding to each of the curves to be detected is compared with the burr threshold to obtain the burr comparison result corresponding to each of the curves to be detected. Based on the stuttering comparison result and the glitch comparison result corresponding to each of the curves to be detected, the detection result corresponding to each of the curves to be detected is obtained.

2. The method according to claim 1, characterized in that, The detection data includes the displacement value and / or thrust value of the cell being pushed into the battery casing; The step of generating one or more curves to be detected based on the detection data includes: Generate a detection curve corresponding to the thrust value based on the thrust value, and / or generate a detection curve corresponding to the displacement value based on the displacement value.

3. The method according to claim 1, characterized in that, The step of detecting one or more curves to be detected based on historical detection data to obtain a detection result for each curve to be detected includes: A standard curve is generated for each curve to be detected based on the historical detection data, and a slope threshold is determined for each standard curve based on each standard curve. The slope of each of the curves to be detected is determined, and the slope of each curve to be detected is compared with the corresponding slope threshold to obtain the slope comparison result of each curve to be detected. Determine the similarity between each of the curves to be detected and the corresponding standard curve, and compare each similarity with a similarity threshold to obtain the similarity comparison result of each curve to be detected; Based on the slope comparison results and similarity comparison results of each of the curves to be detected, the detection result corresponding to each curve to be detected is obtained.

4. The method according to claim 1, characterized in that, The method further includes: Based on the stuttering comparison results corresponding to the target detection curve, a first number of target detection curves with stuttering values ​​greater than the stuttering threshold are determined; the target detection curve is any of the detection curves. Based on the burr comparison results corresponding to the target detection curve, a second number of burr values ​​in the target detection curve that are greater than the burr threshold is determined; Based on the first quantity, the weight of the stuttering comparison result, the second quantity, and the weight of the glitch comparison result, the anomaly score corresponding to the target detection curve is determined; wherein, the weight of the stuttering comparison result is greater than the weight of the glitch comparison result. The abnormality level of the battery cell casing is determined based on the abnormality score corresponding to each of the curves to be detected.

5. The method according to any one of claims 1 to 4, characterized in that, After generating one or more curves to be detected based on the detection data, the method further includes: According to the battery cell assembly process, each of the curves to be tested is divided into multiple sub-curves; each sub-curve corresponds one-to-one with a sub-process of the battery cell assembly process. The step of detecting one or more curves to be detected based on historical detection data to obtain a detection result for each curve to be detected includes: Each sub-curve of each curve to be detected is detected according to historical detection data to obtain the detection result corresponding to each sub-curve of each curve to be detected, and the detection result corresponding to each curve to be detected is determined according to the detection result corresponding to each sub-curve.

6. An abnormality detection device for battery cell insertion, characterized in that, The device includes: The data acquisition module is used to acquire detection data of the battery cell during the process of being pushed into the battery casing. A curve construction module is used to generate one or more curves to be detected based on the detection data; The data detection module is used to determine the abnormal data corresponding to each of the curves to be detected based on the detection data. The curve detection module is used to detect one or more curves to be detected based on historical detection data, and to obtain the detection result corresponding to each curve to be detected; the historical detection data refers to the detection data of the battery cells when no abnormality occurred during the process of the battery cells being pushed into the battery casing. An anomaly detection module is used to determine the abnormal result of the cell insertion into the casing based on the detection result corresponding to each of the curves to be detected. The abnormal data includes stuttering values ​​and glitches; The curve detection module is also used for: Anomaly data thresholds are determined based on historical detection data, including stuttering thresholds and glitch thresholds. The stuttering value corresponding to each of the curves to be detected is compared with the stuttering threshold to obtain the stuttering comparison result corresponding to each of the curves to be detected. The burr value corresponding to each of the curves to be detected is compared with the burr threshold to obtain the burr comparison result corresponding to each of the curves to be detected. Based on the stuttering comparison result and the glitch comparison result corresponding to each of the curves to be detected, the detection result corresponding to each of the curves to be detected is obtained.

7. An electronic device, characterized in that, The system includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images