A lithium battery defect detection system and method based on a PLC control system

Abstract

The application discloses a lithium battery defect detection system and method based on a PLC control system, relates to the technical field of lithium battery automation equipment, and solves the problem that simple image recognition cannot meet the actual production demand, the types and sizes of lithium batteries are different, different detection accuracies are different, and when production demand frequently changes, developing an automatic and flexible lithium battery defect detection system becomes an important demand problem in the industry; the application analyzes surface data and structure data of the lithium battery through image processing, divides the types according to the image paving surface area and structure volume, generates exclusive defect detection training data sets for each type in the subsequent stage, extracts features of the defect detection training data sets by using a detection model based on a deep learning algorithm, classifies defects, and analyzes accuracy, so that targeted defect detection can be performed on lithium batteries of different specifications, and adaptive adjustment of detection standard parameters can be achieved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of lithium battery automation equipment, and particularly relates to a lithium battery defect detection system and method based on a PLC control system. BACKGROUND

[0002] With the wide application of lithium batteries in various portable electronic devices, electric vehicles and other fields, the safety problem of lithium batteries is increasingly prominent. In the production process of lithium batteries, various defects such as electrode misplacement, electrode sheet breakage and electrolyte leakage may be caused due to imperfect process or raw material problems, which will seriously affect the performance and safety of the battery. The traditional defect detection method often relies on image processing technology to train an image recognition model to detect the target of the lithium battery;

[0003] However, pure image recognition cannot meet the actual production needs, and the types and sizes of lithium batteries are different, and there are differences in detection accuracy. When the production demand frequently changes, it is an important requirement of the industry to develop an automatic and flexible parameter-adjustable lithium battery defect detection system. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art; for this purpose, the present application proposes a lithium battery defect detection method based on a PLC control system, which is used to solve the technical problems that it is difficult to ensure stable tracking of the target object, the relative position of the target object changes greatly in a short time, and it is difficult to ensure accurate monitoring of the pose data of the target object.

[0005] To solve the above problems, the first aspect of the present application provides a lithium battery defect detection method based on a PLC control system, comprising the following steps:

[0006] S1. Acquire real-time images of lithium batteries on the production line, generate surface and structure information of the lithium batteries, and on this basis, classify the lithium batteries by model, establish an initial data set including plane data and structure data, and network search the temperature rise curve of the lithium battery overheating, analyze the acceleration of temperature rise, and establish a temperature abnormal data set;

[0007] S2. For each folder Lj, analyze the initial data set of the lithium battery of the corresponding model of the folder Lj, construct a defect detection training data set including size accuracy detection data package, standard data package and defect data package; based on the temperature abnormal data set, construct a temperature training data set;

[0008] S3. Establish an identification model and a comparison model based on a deep algorithm, train the identification model using the defect detection training set, execute image processing and defect recognition algorithm; train the comparison model using the temperature training data set, and execute dynamic time warping algorithm;

[0009] S4. Comprehensive analysis of the monitoring results of the identification model and the comparison model, interface with the PLC control system, generate control instructions, and eliminate unqualified lithium batteries.

[0010] As a further scheme of the present application: the step S1 of establishing the initial data set and the temperature anomaly data set comprises the following steps:

[0011] The establishment step of the initial data set is as follows:

[0012] A plurality of groups of high-speed cameras are set, each group of cameras adopts a plurality of different angles to acquire image information of the lithium battery in real time, the image information is spliced to generate a plan view of the lithium battery;

[0013] Coordinate axis processing is performed on the basis of the plan view, and a plurality of region intervals are generated, denoted as Qi

x, y

[0014] Based on the planar data, a three-dimensional reconstruction technology of multi-view images is used to generate a three-dimensional structure diagram of the lithium battery, i.e., structure data of the lithium battery; each region interval Qi

x, y

[0015] Based on the planar data and the structure data in the initial data set, lithium batteries of different planar area sizes and structure size differences are classified into different models, denoted as Lj, wherein j=1, 2, 3, …, i.e., the model number of the lithium battery, and the planar data and the structure data of lithium batteries of the same model are placed in a folder, and the folder is named according to the model number of the lithium battery;

[0016] The establishment step of the temperature anomaly data set is as follows:

[0017] The temperature rise curve of the lithium battery overheating is searched on the network, the acceleration of the temperature rise is analyzed, and a temperature rise acceleration curve is drawn;

[0018] The starting coordinates of the acceleration in the temperature rise acceleration curve are defined as the mutation origin points when the change is greater than or equal to r within the time R, the mutation origin points are located and marked, the temperature rise acceleration curve of the latter segment of the mutation origin points is analyzed, and the temperature anomaly data set is established.

[0019] As a further scheme of the present application: the step S1 of classifying the lithium batteries into different models comprises the following steps:

[0020] The classification step is: comparing the surface area and volume of the lithium battery, setting a plurality of maximum tolerances according to the comparison result, wherein each group includes a surface area maximum tolerance and a volume maximum tolerance; the data samples covered in each group of tolerances are classified into the same model.

[0021] As a further scheme of the present application: the step S2 of constructing the defect detection training data set and the temperature training data set comprises the following steps:

[0022] For the planar data in each folder, use OPEN CV software to process and generate a gray scale map of the lithium battery surface, analyze the gray scale map, compare the pixel values of each lithium battery in each region interval QI

x, y

[0023] Through the corresponding relationship, the abnormal region is found in the corresponding part of the structural data. Through three-dimensional coordinate operation in the structural data, the average recess depth h3 of the abnormal region perpendicular to the plane, the maximum average recess depth h2, and the thickness h1 of the lithium battery in the vertical direction are calculated, wherein the protruding part is marked as a negative value. Analyze the relationship between h1, h2 and h3 to obtain the complex recess coefficient C complex Draw the complex recess coefficient C complex pixel value correspondence table, and generate a size precision detection data package;

[0024] In each folder, according to the planar data, the abnormal region is cut down and marked as a defect data package, and the other regions in the planar data are marked as normal regions and marked as a standard data package;

[0025] On the basis of the temperature abnormal data set in each folder, real-time current temperature monitoring is set, and the specific operation is as follows:

[0026] By setting the charging and discharging circuit, including the protection circuit, the charging circuit and the output circuit, the lithium battery is connected in series, and the current sensor and the temperature sensor are set to monitor the current and internal temperature changes of the lithium battery in real time and record them. Draw a temperature acceleration real-time curve graph, and mark it as a temperature training data set.

[0027] As a further scheme of the present application: analyze the relationship between h1, h2 and h3 to obtain the complex recess coefficient C complex , comprising the following steps:

[0028] Maximum recess ratio factor:

[0029]

[0030] Average recess ratio factor:

[0031]

[0032] Definition of complex recess coefficient C complex , which represents the influence degree of the recess phenomenon of the abnormal area on the safety production of lithium batteries, the analysis formula is:

[0033]

[0034] Wherein, the symbol of C complex is consistent with R avg and R max , k is an amplification factor, γ is a weight factor, which reflects the importance of the recess shape in the complex recess coefficient C complex , S is a factor describing the recess shape, which is taken as 0.8 here;

[0035] The selection of the amplification factor k depends on the sensitivity requirement of the production line to the recess, and the value is 10; γ is taken as 1.38.

[0036] As a further scheme of the application: according to the plane data, the abnormal area is cut out, including the following steps:

[0037] The edge of the abnormal area is smoothed, a small enough square grid is defined, the edge of the abnormal area is smoothed, the area of the edge of the abnormal area is less than the area of the square grid, the area is completely covered by the minimum number of square grids, so that the length of the projection of the abnormal area in any direction is an integer multiple of the side length of the square grid.

[0038] As a further scheme of the application: the training mode of the step S3, identifying the model, includes the following steps:

[0039] By inputting the standard data package, which includes each type of lithium battery, the characteristics of each type of lithium battery are extracted, and the standard interval of any characteristic is determined; the specific characteristics include the surface area, volume, surface scratch, accompanying foreign matter, swelling degree, electrode misplacement and coating loss of lithium battery;

[0040] By inputting the defect data package, which includes each type of lithium battery, the defects of each type of lithium battery are classified, and for each type of lithium battery, the defect types of each type are divided according to the extracted characteristics, wherein the defects not found corresponding characteristics are recorded as other defects;

[0041] By inputting the size accuracy detection data package, which includes each type of lithium battery, the size accuracy of each type of lithium battery is divided; for the surface scratch and swelling degree of each type of lithium battery, the preset boundary value of the complex recess coefficient C complex is set, and the setting method is to take the complex recess coefficient C complexIn the pixel value correspondence table, the complex concave coefficient value corresponding to the lowest pixel value of the abnormal area is a preset boundary value;

[0042] If the complex concave coefficient C of the surface scratch exceeds the corresponding preset boundary value, it is defined as unqualified in size accuracy; otherwise, it is defined as qualified in size accuracy. complex If the complex concave coefficient C of the surface scratch exceeds the corresponding preset boundary value, it is defined as unqualified in size accuracy; otherwise, it is defined as qualified in size accuracy.

[0043] If the complex concave coefficient C of the expansion degree exceeds the corresponding preset boundary value, it is defined as unqualified in size accuracy; otherwise, it is defined as qualified in size accuracy. complex If the complex concave coefficient C of the expansion degree exceeds the corresponding preset boundary value, it is defined as unqualified in size accuracy; otherwise, it is defined as qualified in size accuracy.

[0044] As another aspect of the present application, the step S3, the training method of the comparison model, includes the following steps:

[0045] The temperature training data set is input, which includes each type of lithium battery. For each type of lithium battery, the feature extraction of the curve after the mutation origin of the temperature rising acceleration curve is carried out, including the overall rising trend and the segmented rising trend. The overall rising trend refers to the curve shape after the mutation origin, and the segmented rising trend refers to the curve shape of the overall rising trend in a certain time interval, and the longest time interval is not more than one second.

[0046] As another aspect of the present application, the step S4, the interface PLC control system generates control instructions to remove unqualified lithium batteries, including the following steps:

[0047] The identification model and the comparison model are linked to the PLC control system, and the type of the lithium battery is determined by its volume and surface area. If any of the features of the accompanying foreign matter, electrode misplacement and coating loss is detected, the coordinated removal device intercepts the target lithium battery.

[0048] If other defects are detected, the coordinated removal device intercepts the target lithium battery.

[0049] For the expansion degree of any type of lithium battery, if it is divided into unqualified, the coordinated removal device intercepts the target lithium battery; otherwise, it does not act.

[0050] For the surface scratch of any type of lithium battery, if it is divided into unqualified, the coordinated removal device intercepts the target lithium battery; otherwise, it does not act.

[0051] The lithium batteries are connected in series to the charging and discharging circuit, and S1, S2 and S3 are repeated to monitor the expansion degree and temperature rising acceleration of the lithium batteries in the charging and discharging circuit:

[0052] If the similarity between the temperature acceleration rising trend of the lithium battery and the overall rising trend is greater than or equal to 40%, the lithium battery is marked as unqualified, and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken;

[0053] If the temperature acceleration of the lithium battery has at least one rising trend, and there are two or more consecutive segmented rising trends with a similarity greater than 50%, the lithium battery is marked as unqualified, and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken;

[0054] If the temperature acceleration rising trend of the lithium battery and the segmented rising trend have several discontinuous segmented rising trends, and the similarity of each segment is greater than or equal to 30%, the complex depression coefficient C of the lithium battery is comprehensively analyzed. complex ; Set the secondary preset boundary value, which is derived from the complex depression coefficient C of the qualified lithium battery when it expands during the charge and discharge process. complex The average value of the complex depression coefficient C of the lithium battery in the segmented rising trend time range with a similarity greater than or equal to 30%. complex If the value is greater than the mean, the lithium battery is deemed unqualified and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken;

[0055] As another aspect of the present invention, a lithium battery defect detection system based on a PLC control system is provided. The system specifically adopts a lithium battery defect detection method based on a PLC control system to detect lithium battery defects.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The present invention analyzes the surface and structural data of lithium batteries through image processing, and divides lithium batteries into models of different specifications based on the surface area and structural volume of the image tile. Subsequently, a dedicated defect detection training data set is generated for each model. The detection model based on the deep learning algorithm is used to perform feature extraction, defect classification, and accuracy analysis on the defect detection training data set, thereby achieving targeted defect detection for lithium batteries of different specifications and adaptive adjustment of detection standard parameters.

[0058] Analyze the temperature rise acceleration curve of abnormal lithium batteries under temperature out-of-control conditions. Through the trend of the real-time temperature rise acceleration curve of the lithium battery, the consequences of abnormal lithium battery temperature can be predicted in time, effectively avoiding production accidents.

[0059] The volume gradient and the corresponding surface area gradient of the lithium battery are simply divided, the storage power capacity of the lithium battery is divided, the model is divided according to the actual situation, and subsequent analysis is carried out on the lithium battery of different models; the requirements of different types of lithium batteries on precision, the types and severity of defects in the production process are made into automatic and more flexible detection standards;

[0060] The complex recess coefficient conforming to the standard is set in advance, the expansion of the lithium battery caused by various elements in the production process is detected by the more accurate standard, whether the lithium battery reaches the qualified production quality for use, whether the size of the lithium battery reaches the standard, and the fault tolerance range limit and limit index are provided for actual production;

[0061] The abnormal area is edged, the data processing and subsequent analysis are facilitated, the abnormal area is cropped, the coordinate interval of the abnormal area is better positioned, the abnormal area is covered with a simplified square grid, the difficulty of image edge processing is optimized, and the minimum number of square grids greatly approximates the actual abnormal area, and the difference between the simplified model and the actual value is reduced;

[0062] The present application detects the internal circuit of the lithium battery, analyzes the temperature change trend of the lithium battery, diagnoses the state of the lithium battery, and excludes abnormal lithium batteries. For hidden dangers that cannot be directly judged from the appearance, the temperature change of the lithium battery and the similarity of abnormal cases in the charging and discharging process are better and faster to exclude unqualified lithium batteries, and timely exclude. BRIEF DESCRIPTION OF DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0064] Figure 1 The method flowchart of the present application is shown in the figure;

[0065] Figure 2 The system framework of the present application is shown in the figure. DETAILED DESCRIPTION

[0066] The technical solutions of the present application will be described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0067] Embodiment one

[0068] Referring to Figure 1 The first aspect of the present application provides a lithium battery defect detection method based on a PLC control system, comprising the following steps:

[0069] S1. Obtain real-time images of lithium batteries on the production line to generate surface and structure information of the lithium batteries, and on this basis, classify the lithium batteries by model, establish an initial data set, including plane data and structure data; network search for temperature rise curves of lithium batteries overheating, analyze the acceleration of temperature rise, and establish a temperature anomaly data set;

[0070] S2. For each folder Lj, analyze the initial data set of lithium batteries of the corresponding model of the folder Lj, construct a defect detection training data set, including size precision detection data packets, standard data packets and defect data packets; based on the temperature anomaly data set, construct a temperature training data set;

[0071] S3. Establish a recognition model and a comparison model based on a deep algorithm, train the recognition model using the defect detection training set, and execute image processing and defect recognition algorithms; train the comparison model using the temperature training data set, and execute a dynamic time warping algorithm;

[0072] S4. Comprehensive analysis of the monitoring results of the recognition model and the comparison model, interface with the PLC control system, generate control instructions, and eliminate unqualified lithium batteries.

[0073] In one embodiment of the present application, the step S1 of establishing the initial data set and the temperature anomaly data set comprises the following steps:

[0074] The establishment steps of the initial data set are as follows:

[0075] A plurality of high-speed cameras are set up, each group of cameras uses multiple different angles to obtain real-time image information of the lithium batteries, and the image information is spliced to generate a plan view of the lithium batteries;

[0076] On the basis of the plan view, coordinate axis processing is performed, and a plurality of area intervals are generated, denoted as Qi

x, y

[0077] Based on the planar data, a three-dimensional reconstruction technology of multi-view images is used to generate a three-dimensional structure diagram of the lithium battery, i.e. the structure data of the lithium battery; each area interval Qi

x, y

[0078] Based on the planar data and structural data in the initial data set, for different planar area sizes and structural size differences, lithium batteries are classified by model, denoted as Lj, wherein j=1, 2, 3,..., that is, the model number of the lithium battery, and the planar data and structural data of lithium batteries of the same model are placed in a folder, and the folder is named according to the lithium battery model;

[0079] The temperature anomaly data set is established as follows:

[0080] The temperature rise curve of the lithium battery overheating is searched on the network, and the acceleration of the temperature rise is analyzed to draw the temperature rise acceleration curve.

[0081] The starting coordinates of the acceleration in the temperature rise acceleration curve are defined as the mutation origin points when the change is greater than or equal to r within time R, the mutation origin points are located and marked, and the temperature rise acceleration curve of the latter section of the mutation origin points is analyzed to establish the temperature anomaly data set.

[0082] By simply dividing the volume gradient and the corresponding surface area gradient of the lithium battery, the storage power capacity of the lithium battery is divided, the model is divided according to the actual situation, and subsequent analysis is carried out for lithium batteries of different models; for different types of lithium batteries, the requirements for precision, the types and severity of defects in the production process are made into automatic, more flexible detection standards, such as large power batteries may pay more attention to the alignment of the electrodes and the leakage problem of the electrolyte, while small portable batteries may focus more on surface damage and size precision.

[0083] Analyzing the temperature rise acceleration curve of the abnormal lithium battery under the condition of temperature out of control can predict the consequences of lithium battery temperature anomaly in time through the trend of the real-time temperature rise acceleration curve of the lithium battery, and avoid production accidents.

[0084] In one embodiment of the present application, the step S1 of classifying the lithium battery by model comprises the following steps:

[0085] The classification step is: comparing the surface area and volume of the lithium battery, setting a plurality of groups of maximum tolerances according to the comparison results, wherein each group includes a surface area maximum tolerance and a volume maximum tolerance; the data samples covered in each group of tolerances are classified into the same model.

[0086] In one embodiment of the present application, the step S2 of constructing the defect detection training data set and the temperature training data set comprises the following steps:

[0087] For the planar data in each folder, use OPEN CV software to process, generate a grayscale image of the lithium battery surface, analyze the grayscale image, compare the pixel value of each lithium battery in each area interval QI

x, y

[0088] Through the corresponding relationship, the abnormal area is found in the corresponding part of the structure data. In the structure data, the average recess depth h3, the maximum average recess depth h2 and the thickness h1 of the lithium battery in the vertical direction are calculated by three-dimensional coordinate operation, wherein the protruding part is negative. Analyze the relationship between h1, h2 and h3 to obtain the complex recess coefficient C complex , draw the complex recess coefficient C complex - pixel value corresponding table, and generate a size precision detection data package;

[0089] The complex recess coefficient conforming to the standard is set in advance. For the expansion and slight deformation of the lithium battery caused by various elements in the production process, the production quality of the lithium battery reaching the qualified use is detected by more accurate standard, and whether the size of the lithium battery reaches the standard is better detected, so as to provide a certain fault tolerance range limit and limitation index for actual production;

[0090] In each folder, the abnormal area is cut down according to the planar data, and is marked as a defect data package. Other areas in the planar data are marked as normal areas, and are marked as a standard data package;

[0091] On the basis of the temperature abnormal data set in each folder, real-time current temperature monitoring is set, and the specific operation is as follows:

[0092] By setting the charging and discharging circuit including the protection circuit, the charging circuit and the output circuit, the lithium battery is connected in series, and the current sensor and the temperature sensor are set, the current and internal temperature change of the lithium battery are monitored and recorded in real time, and the temperature acceleration real-time curve is drawn, which is marked as a temperature training data set;

[0093] The present application analyzes the change trend of the temperature of the lithium battery by detecting the internal circuit of the lithium battery, diagnoses the state of the lithium battery, and excludes abnormal lithium batteries. For hidden dangers that cannot be directly judged from the appearance, the temperature change of the lithium battery and the similarity of abnormal cases in the charging and discharging process are better and faster to exclude unqualified lithium batteries and timely exclude.

[0094] In one embodiment of the present application, the relationship between h1, h2 and h3 is analyzed to obtain the complex recess coefficient C complex , comprising the following steps:

[0095] The maximum recess ratio factor is:

[0096]

[0097] Average recess ratio factor:

[0098]

[0099] Define complex recess coefficient C complex , which represents the influence degree of the recess phenomenon of the abnormal area on the safety production of lithium batteries, and the analysis formula is:

[0100]

[0101] Wherein, C complex The sign of R avg and R max is consistent, k is the amplification factor, γ is the weight factor, the importance of the recess shape in the complex recess coefficient C complex , S is the factor describing the recess shape, which is 0.8 here;

[0102] The selection of the amplification factor k depends on the sensitivity requirement of the production line to the recess, and the value is 10; γ is 1.38.

[0103] In one embodiment of the present application, the abnormal area is cut out according to the plane data, including the following steps:

[0104] The edge of the abnormal area is smoothed, a small enough square grid is defined, the edge of the abnormal area is smoothed, the area of the edge of the abnormal area is less than the area of the square grid, and the edge of the abnormal area is completely covered by the minimum number of square grids, so that the length of the projection of the abnormal area in any direction is an integer multiple of the side length of the square grid;

[0105] The edge of the abnormal area is processed, which facilitates the processing and subsequent analysis of the data, the cutting of the abnormal area, the better positioning of the coordinate interval of the abnormal area, the covering of the abnormal area with a simplified square grid, and the optimization of the difficulty of image edge processing, and the minimum number of square grids greatly approximates the actual abnormal area, reducing the difference between the simplified model and the actual value.

[0106] In one embodiment of the present application, the step S3, the training mode of the model, includes the following steps:

[0107] By inputting the standard data package, which includes the lithium battery of each model, the characteristics of each model of lithium battery are extracted, and the standard interval of any characteristic is determined; the specific characteristics include the surface area, volume, surface scratch, accompanying foreign matter, expansion degree, electrode misplacement and coating loss of lithium battery;

[0108] By inputting the defect data package including each type of lithium battery, the defect classification of each type of lithium battery is carried out, and for each type of lithium battery, the defect type is divided according to the extracted features, wherein the defect not found corresponding feature is recorded as other defects;

[0109] By inputting the size precision detection data package including each type of lithium battery, the size precision of each type of lithium battery is divided; for the surface scratch and the expansion degree of each type of lithium battery, the preset boundary value of the complex indentation coefficient C complex of the type of lithium battery is set, and the setting mode is to take the complex indentation coefficient C complex of the type of lithium battery. In the pixel value corresponding relationship table, the complex indentation coefficient value corresponding to the lowest pixel value of the abnormal area is the preset boundary value.

[0110] If the complex indentation coefficient C complex of the surface scratch exceeds the corresponding preset boundary value, the size precision is defined as unqualified; otherwise, the size precision is defined as qualified.

[0111] If the complex indentation coefficient C complex of the expansion degree exceeds the corresponding preset boundary value, the size precision is defined as unqualified; otherwise, the size precision is defined as qualified.

[0112] In one embodiment of the present application, the training mode of the comparison model includes the following steps:

[0113] By inputting the temperature training data set including each type of lithium battery, the feature extraction of the curve mutation original point after section of the temperature rising acceleration curve of each type of lithium battery is carried out, including the overall rising trend and the segmented rising trend, wherein the overall rising trend refers to the curve shape after the mutation original point, and the segmented rising trend is the curve shape of the overall rising trend in a certain time interval, and the longest time interval is not more than one second.

[0114] In one embodiment of the present application, the step S4, the interface PLC control system generates the control instruction to remove the unqualified lithium battery, including the following steps:

[0115] The identification model and the comparison model are linked to the PLC control system, the type of the lithium battery is determined by the volume and the surface area of the lithium battery, and if any feature of the accompanying foreign matter, electrode misplacement and coating loss is detected, the coordination removal device is coordinated to intercept the target lithium battery;

[0116] If other defects are detected, the coordination removal device is coordinated to intercept the target lithium battery.

[0117] For the expansion degree of any type of lithium battery, if it is divided into unqualified, that is, the removal device is coordinated, the target lithium battery is intercepted; otherwise, no action;

[0118] For the surface scratch of any type of lithium battery, if it is divided into unqualified, that is, the removal device is coordinated, the target lithium battery is intercepted; otherwise, no action;

[0119] Connect the lithium battery in series into the charging and discharging circuit, and repeat S1, S2 and S3 to monitor the expansion degree and temperature rising acceleration of the lithium battery in the full circuit:

[0120] If the temperature acceleration rising trend of the lithium battery is similar to the overall rising trend, the similarity is greater than or equal to 40%, the lithium battery is recorded as unqualified, that is, the removal device is coordinated, the target lithium battery is intercepted; otherwise, no action;

[0121] If the temperature acceleration rising trend of the lithium battery has at least one place, the similarity of two or more continuous segmented rising trends is greater than 50%, the lithium battery is recorded as unqualified, that is, the removal device is coordinated, the target lithium battery is intercepted; otherwise, no action;

[0122] If the temperature acceleration rising trend of the lithium battery has several discontinuous segmented rising trends, and the similarity of each place is greater than or equal to 30%, the complex concave coefficient C complex of the lithium battery is analyzed comprehensively; a secondary preset boundary value is set, which is derived from the average value of the complex concave coefficient C complex of the qualified lithium battery when it expands during the charging and discharging process; if the complex concave coefficient C complex of the lithium battery is greater than the average value in the segmented rising trend time range with a similarity greater than or equal to 30%, the lithium battery is recorded as unqualified, that is, the removal device is coordinated, the target lithium battery is intercepted; otherwise, no action.

[0123] Embodiment two

[0124] In another aspect of the present application, please refer to Figure 2 A lithium battery defect detection system based on a PLC control system, which specifically adopts a lithium battery defect detection method based on a PLC control system to detect the defects of the lithium battery:

[0125] Data acquisition unit: acquire real-time images of lithium batteries on the production line, generate surface and structure information of lithium batteries, classify lithium batteries by type on this basis, establish an initial data set including plane data and structure data; network search temperature rising curve of lithium battery overheating, analyze temperature rising acceleration, and establish temperature anomaly data set;

[0126] Data integration unit: for each folder Lj, analyze the initial data set of the lithium battery of the model corresponding to the folder Lj, construct a defect detection training data set, including a size precision detection data package, a standard data package and a defect data package; based on the temperature abnormal data set, construct a temperature training data set;

[0127] Model training unit: establish an identification model based on a deep algorithm, train the identification model by using the defect detection training set; train the comparison model by using the temperature training data set;

[0128] Detection unit: based on the detection model, execute image processing and defect recognition algorithm to obtain a prediction result; execute dynamic time warping algorithm according to the comparison model to generate a detection result;

[0129] PLC control unit: comprehensively analyze the monitoring results of the identification model and the comparison model, interface with a PLC control system, generate a control instruction, and eliminate unqualified lithium batteries.

[0130] Embodiment three

[0131] As embodiment three of the present application, the technical scheme of the embodiment is to combine the schemes of the two preceding embodiments.

[0132] The above embodiments are only used to illustrate the technical method of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A lithium battery defect detection method based on a PLC control system, characterized in that: The following steps are involved: S1. Acquire real-time images of lithium batteries on the production line, generate surface and structural information, classify the batteries by type, and establish an initial dataset, including surface and structural data. Retrieve the temperature rise curve of lithium battery overheating online, analyze the acceleration of temperature rise, and establish a temperature anomaly data set; S2. For each folder Lj, analyze the initial data set of lithium batteries corresponding to the model of the folder Lj to construct a defect detection training data set, including dimensional accuracy detection data packets, standard data packets, and defect data packets; Based on the temperature anomaly dataset, a temperature training dataset is constructed; S3. Establish a recognition model and a comparison model based on a deep learning algorithm. Use the defect detection training set to train the recognition model and execute the image processing and defect recognition algorithms. Use the temperature training dataset to train the comparison model and execute the dynamic time warping algorithm. S4. Comprehensively analyze the monitoring results of the identification model and the comparison model, connect to the PLC control system, generate control instructions, and eliminate unqualified lithium batteries; The step S1, establishing an initial data set and a temperature anomaly data set, includes the following steps: The steps to build the initial dataset are as follows: Set up several groups of high-speed cameras, each group of cameras uses multiple different angles to obtain image information of the lithium battery in real time, splice the image information, and generate a plan view of the lithium battery; Based on the plane graph, coordinate axis processing is performed to generate several area intervals, recorded as Qi [x, y], where x is the horizontal coordinate of the coordinate axis, y is the vertical coordinate of the coordinate axis, and i = 1, 2, 3, ...; forming the plane data of the lithium battery; Based on the plane data, a three-dimensional structure diagram of the lithium battery, i.e., the structural data of the lithium battery, is generated through multi-view image 3D reconstruction technology; each region interval Qi[x, y] corresponds to a plane view of the three-dimensional structure diagram of the lithium battery, and the coordinate axes of the plane view on the three-dimensional structure diagram are made to coincide with the coordinate axes on the plane diagram; Based on the plane data and structural data in the initial dataset, lithium batteries are classified into different models according to different plane area sizes and structural size differences, denoted as Lj, where j = 1, 2, 3, etc., which is the lithium battery model number. The plane data and structural data of lithium batteries of the same model are placed in a folder, and the folder is named after the lithium battery model. The steps to build the temperature anomaly dataset are as follows: Retrieve the temperature rise curve of lithium battery overheating online, analyze the acceleration of temperature rise, and draw the temperature rise acceleration curve; In the temperature rise acceleration curve, the starting coordinate where the acceleration changes by more than or equal to r within the time R is defined as the mutation origin. The mutation origin is located and marked, and the temperature rise acceleration curve after the mutation origin is analyzed to establish a temperature anomaly dataset.

2. The method for detecting defects in lithium batteries based on a PLC control system according to claim 1, wherein: The step S1, classifying the lithium batteries by model, comprises the following steps: The classification steps are as follows: compare the surface area and volume of the lithium battery, and set several groups of maximum tolerances based on the comparison results, where each group includes the maximum tolerance for the surface area and the maximum tolerance for the volume; the data samples covered within each group of tolerances are classified as the same model.

3. The lithium battery defect detection method based on a PLC control system according to claim 1, characterized in that: The step S2, constructing a defect detection training data set and a temperature training data set, includes the following steps: For the plane data in each folder, use OPEN CV software to process it and generate a grayscale image of the lithium battery surface. Analyze the grayscale image and compare the pixel values ​​of each lithium battery in each area interval Qi [x, y]. Mark all areas with different pixel values ​​and a small number of samples as abnormal areas. Through the corresponding relationship, the abnormal area is found in the corresponding part of the structural data. In the structural data, the average depression depth h3, the maximum average depression depth h2, and the thickness h1 of the lithium battery in the vertical direction of the abnormal area are calculated through three-dimensional coordinate calculation. Among them, the protruding part is recorded as a negative value; the relationship between h1, h2, and h3 is analyzed to obtain the complex depression coefficient , plot the complex concavity coefficient -Pixel value correspondence table and generate size accuracy detection data packet; In each folder, based on the plane data, the abnormal area is cut out and recorded as a defective data packet, and the other areas in the plane data are recorded as normal areas and recorded as standard data packets; Based on the temperature anomaly dataset in each folder, set up real-time current and temperature monitoring. The specific operations are as follows: By setting up a charging and discharging circuit, including a protection circuit, a charging circuit, and an output circuit, the lithium battery is connected in series, and a current sensor and a temperature sensor are set up to monitor and record the lithium battery current and internal temperature changes in real time, draw a real-time temperature acceleration curve, and record it as a temperature training data set.

4. The method for detecting defects in a lithium battery based on a PLC control system according to claim 3, wherein: Analyze the relationship between h1, h2, and h3 to obtain the complex concave coefficient , including the following steps: Maximum sag scale factor: ； Average Sag Scale Factor: ； Define complex concavity coefficient , which indicates the degree of influence of the depression phenomenon in the abnormal area on the safe production of lithium batteries. The analysis formula is: ； in, The symbol and and Keep it consistent, k is the amplification factor, is the weight factor, reflecting the concave shape in the complex concave coefficient The importance of ist, S is a factor describing the shape of the concave.

5. The method for detecting defects in lithium batteries based on a PLC control system according to claim 3, wherein: Based on the plane data, the abnormal area is clipped, including the following steps: Fill the abnormal area, define a sufficiently small square grid, and smooth the edges of the abnormal area so that the area where the edge of the abnormal area is smaller than the area of ​​the square grid is completely covered by the minimum number of square grids, so that the length of the projection of the abnormal area in any direction is an integer multiple of the side length of the square grid.

6. The method for detecting defects in lithium batteries based on a PLC control system according to claim 1, wherein: The step S3, identifying the training mode of the model, comprises the following steps: By inputting a standard data package that includes each model of lithium battery, feature extraction is performed on each model of lithium battery to determine the standard range of any feature; specific features include lithium battery surface area, volume, surface scratches, accompanying foreign matter, expansion degree, electrode misalignment and coating loss; By inputting a defect data packet, which includes each model of lithium battery, each model of lithium battery is defect-classified. For each model of lithium battery, the defect type of each model is divided according to the extracted features. Among them, defects for which no corresponding features are found are recorded as other defects; By inputting the dimensional accuracy detection data package, which includes each model of lithium battery, the dimensional accuracy of each model of lithium battery is divided; according to the surface scratches and expansion degree of each model of lithium battery, the complex concave coefficient is set separately The preset boundary value is set by taking the complex concavity coefficient of this type of lithium battery - In the pixel value correspondence table, the complex concave coefficient value corresponding to the lowest pixel value in the abnormal area is the preset boundary value; If the complex concave coefficient of the surface scratch If the corresponding preset boundary value is exceeded, the dimensional accuracy is defined as unqualified; otherwise, the dimensional accuracy is defined as qualified; If the complex concavity coefficient of the expansion degree If the value exceeds the corresponding preset boundary value, the dimensional accuracy is defined as unqualified; otherwise, the dimensional accuracy is defined as qualified.

7. The method for detecting defects in lithium batteries based on a PLC control system according to claim 1, wherein: The step S3, comparing the training method of the model, includes the following steps: By inputting the temperature training data set, which includes each model of lithium battery, the feature extraction of the curve after the mutation origin of the temperature rise acceleration curve is performed for each model of lithium battery, including the overall rising trend and the segmented rising trend. Among them, the overall rising trend refers to the curve shape after the mutation origin, and the segmented rising trend is the curve shape of the overall rising trend within a certain time interval, and the longest time interval does not exceed one second.

8. The method for detecting defects in lithium batteries based on a PLC control system according to claim 1, wherein: The step S4, connecting to the PLC control system, generating control instructions, and removing unqualified lithium batteries, includes the following steps: The identification model and the comparison model are linked to the PLC control system to determine the model of the lithium battery based on its volume and surface area. If any of the following features, such as foreign matter, electrode misalignment, and coating loss, are detected, the rejection device is coordinated to intercept the target lithium battery. If other defects are detected, coordinate the rejection device to intercept the target lithium battery; For any type of lithium battery, if the expansion degree is classified as unqualified, the rejection device will be coordinated to intercept the target lithium battery; otherwise, no action will be taken; For any type of lithium battery with surface scratches, if it is classified as unqualified, the rejection device will be coordinated to intercept the target lithium battery; otherwise, no action will be taken; Connect the lithium battery in series to the charge and discharge circuit, and repeat S1, S2 and S3 to monitor the expansion degree and temperature rise acceleration of the lithium battery in the full circuit: If the similarity between the temperature acceleration rising trend of the lithium battery and the overall rising trend is greater than or equal to 40%, the lithium battery is marked as unqualified, and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken; If the temperature acceleration of the lithium battery increases in at least one segment, and the similarity of two or more consecutive segmented increasing trends is greater than 50%, the lithium battery is marked as unqualified, and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken; If the temperature acceleration rising trend of the lithium battery and the segmented rising trend have several discontinuous segmented rising trends, and the similarity of each segment is greater than or equal to 30%, the complex depression coefficient of the lithium battery is comprehensively analyzed. ; Set the secondary preset boundary value, which is derived from the complex depression coefficient of the qualified lithium battery when it expands during the charge and discharge process The average value of the complex concave coefficient of the lithium battery in the segmented rising trend time range with a similarity greater than or equal to 30% If the value is greater than the mean, the lithium battery is deemed unqualified, and the rejection device is coordinated to intercept the target lithium battery; otherwise, no action is taken.

9. A lithium battery defect detection system based on a PLC control system, wherein the system specifically adopts the method according to any one of claims 1 to 8 to detect lithium battery defects.

Images