High-precision battery pack coating operation method and system based on cylinder servo components

Abstract

This invention discloses a high-precision battery coating operation method and system based on a cylinder servo assembly, relating to the field of coating technology. The method includes: receiving coating requirement parameters of the target battery when it is transported to a preset coating station; optimizing coating control parameters based on the coating requirement parameters to generate target coating control parameters; inputting the target coating control parameters into a servo cylinder control platform of the cylinder servo assembly for control, and acquiring real-time image data of the film bonding state through a vision inspection device; inputting the film bonding state image data into a pre-trained film bonding detection model for analysis, and outputting bonding anomaly detection results; and adjusting the target coating control parameters based on the bonding anomaly detection results. This solves the technical problems of unstable coating quality and low efficiency in the battery coating process of existing technologies, achieving the technical effect of improving coating quality stability and production efficiency.

Description

Technical Field

[0001] This invention relates to the field of coating technology, and more specifically to a high-precision battery coating operation method and system based on a cylinder servo component. Background Technology

[0002] Existing battery coating processes are widely used in the production of various batteries. Their main purpose is to provide an additional protective layer to enhance the battery's mechanical strength, prevent damage from the external environment, and improve the battery's appearance consistency. However, in traditional battery coating technologies, the stability of the coating quality is often affected by multiple factors, such as improper tension control of the film material, film placement deviations, and insufficient mechanical precision of the coating equipment. These factors can lead to quality defects such as wrinkles, bubbles, and film misalignment during the coating process, affecting the overall performance and appearance of the battery. Furthermore, traditional coating processes often rely on manual monitoring and adjustment of coating parameters, resulting in low production efficiency and difficulty in meeting the requirements of high precision and consistency in large-scale production. Although some equipment is equipped with automated control, the lack of high-precision real-time detection methods makes it difficult to detect coating abnormalities in a timely manner and make adjustments, further limiting the improvement of coating quality. Summary of the Invention

[0003] This application provides a high-precision battery packing operation method and system based on a cylinder servo component, which solves the technical problems of unstable film quality and low efficiency in the battery packing process in the prior art.

[0004] In view of the above problems, this application provides a high-precision battery pack coating operation method and system based on a cylinder servo component.

[0005] A first aspect of this application provides a high-precision battery packing operation method based on a cylinder servo assembly, the method comprising:

[0006] When the target battery is delivered to the preset coating station, the coating requirement parameters of the target battery are received. The preset coating station controls the film delivery and coating process via a cylinder servo assembly. Based on the coating requirement parameters, the coating control parameters are optimized to generate target coating control parameters. These target coating control parameters are then input into the servo cylinder control platform of the cylinder servo assembly for control, and a vision inspection device collects real-time image data of the film bonding state. This image data is then input into a pre-trained film bonding detection model for analysis, outputting bonding anomaly detection results. Based on these anomaly detection results, the target coating control parameters are adjusted accordingly.

[0007] A second aspect of this application provides a high-precision battery pack coating system based on a cylinder servo assembly, the system comprising:

[0008] Data receiving module: When the target battery is delivered to the preset coating station, it receives the coating requirement parameters of the target battery, wherein the preset coating station controls the film delivery and coating through a cylinder servo assembly; Optimization module: Based on the coating requirement parameters, it optimizes the coating control parameters to generate target coating control parameters; Control module: It inputs the target coating control parameters into the servo cylinder control platform of the cylinder servo assembly for control, and collects real-time image data of the film bonding status through a vision inspection device; Analysis module: It inputs the image data of the film bonding status into a pre-trained film bonding detection model for analysis and outputs the bonding anomaly detection result; Feedback adjustment module: Based on the bonding anomaly detection result, it adjusts the target coating control parameters.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] When the target battery is conveyed to the preset coating station, its coating requirement parameters are received. The preset coating station controls the film feeding and coating process via a cylinder servo assembly. Next, the coating control parameters are optimized based on the coating requirement parameters to generate target coating control parameters. These target coating control parameters are then input into the servo cylinder control platform of the cylinder servo assembly for control, and real-time image data of the film bonding status is acquired via a vision inspection device. Furthermore, the image data of the film bonding status is input into a pre-trained film bonding detection model for analysis, outputting bonding anomaly detection results. Finally, the target coating control parameters are adjusted based on the bonding anomaly detection results. This solves the technical problems of unstable film bonding quality and low efficiency in the battery coating process of existing technologies, achieving the technical effect of improving coating quality stability and production efficiency. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic flowchart of a high-precision battery pack coating operation method based on a cylinder servo component provided in this application embodiment;

[0013] Figure 2 A schematic diagram of a high-precision battery pack coating system based on a cylinder servo component provided in an embodiment of this application.

[0014] Explanation of reference numerals in the attached diagram: Data receiving module 11, optimization module 12, control module 13, analysis module 14, feedback adjustment module 15. Detailed Implementation

[0015] This application provides a high-precision battery packing operation method and system based on a cylinder servo component, which solves the technical problems of unstable film quality and low efficiency in the battery packing process in the prior art.

[0016] 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 a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown, this application provides a high-precision battery pack coating operation method based on a cylinder servo component, wherein the method includes:

[0019] When the target battery is delivered to the preset coating station, the coating requirement parameters of the target battery are received. The preset coating station controls the film delivery and coating through a cylinder servo component.

[0020] When the target battery is transported to the preset coating station via the transmission device, the identification system (such as an RFID reader or barcode scanner) in the station is first activated to identify the target battery in order to confirm the specific information of the battery and extract the relevant coating requirement parameters. The coating requirement parameters include, but are not limited to, the size specifications of the target battery (length, width, thickness), the type of film material required for coating (e.g., material, thickness, flexibility), the film application accuracy requirements (e.g., positional tolerance range), as well as parameters such as operating speed and film tension.

[0021] The pre-set coating station utilizes an integrated cylinder servo assembly to perform the film delivery and coating operations. First, the servo control system adjusts the initial state of the film delivery mechanism based on received coating requirement parameters, including the initial position of the film, tension control, and delivery speed, ensuring the film moves along the target path to the predetermined position on the battery surface. Subsequently, the servo cylinder assembly executes the film application operation according to the set application path and pressure parameters. The entire process is supported by a real-time feedback mechanism from the servo control platform, which dynamically adjusts the cylinder displacement, speed, and pressure through position sensors and a force control system to ensure stable film application throughout the process, preventing quality issues such as wrinkles, misalignment, or bubbles.

[0022] Furthermore, the preset wrapping station includes a U-shaped wrapping station and a circular wrapping station; wherein, both the U-shaped wrapping station and the circular wrapping station are transported in an equidistant transport mode, and the U-shaped wrapping station is located before the circular wrapping station.

[0023] In the U-shaped wrapping station, the wrapping process employs a U-shaped film application method. The film is applied starting from one side of the battery, sequentially covering the bottom and the other side to complete basic three-sided wrapping. This station uses a cylinder servo assembly to precisely control the film's feeding position, tension, and application force, ensuring the film is flat and wrinkle-free during the wrapping process. The U-shaped wrapping station primarily handles the initial fixing of the battery wrapping, providing a foundation for the subsequent more complex U-shaped wrapping process.

[0024] The loop-shaped wrapping station is responsible for performing a more precise wrapping operation on the battery. Starting from the top of the battery, the membrane wraps around the uncovered parts of the battery (such as the top and two sides) in a loop-shaped path to form a complete 360-degree wrapping effect. This station also uses a cylinder servo assembly and a high-precision control platform for operation. By adjusting the position and tension of the membrane multiple times, the final bonding of the battery is completed, ensuring that the membrane adheres tightly to all contact surfaces.

[0025] Whether at a U-shaped or U-shaped packaging station, the target batteries are handled using an equidistant transport method. Through precise control of conveyor belts or robotic arms, the batteries move to each station at fixed time intervals and distances, preventing disruptions to the packaging rhythm due to positional or time errors. This transport method also facilitates unified control of the overall work cycle, improving production line efficiency.

[0026] Based on the coating requirement parameters, the coating control parameters are optimized to generate the target coating control parameters.

[0027] Based on the received coating requirement parameters, the system optimizes the coating control parameters to generate target coating control parameters that meet the current coating requirements. For example, optimization may involve adjusting membrane tension, application speed, or pressure parameters to ensure that the operation meets the specific coating requirements of the battery.

[0028] Furthermore, based on the coating requirement parameters, the coating control parameters are optimized to generate target coating control parameters, including:

[0029] Establish a coating optimization space; perform digital twin modeling on the preset coating station to establish a twin coating station; randomly select a first coating control parameter in the coating optimization space and input it into the twin coating station for simulation to generate a first coating simulation result; evaluate the coating quality based on the first coating simulation result to generate a first quality index; if the first quality index meets the preset quality constraints, use the first coating control parameter as the target coating control parameter.

[0030] Specifically, an optimization space for coating control parameters is established based on the coating requirement parameters. This optimization space includes the adjustable range of film conveying speed, tension range, pressure range and displacement range of the cylinder servo component, and the film application path parameters. The upper and lower limits and constraints of each control parameter are clearly defined to ensure that all parameter combinations during optimization meet actual operational capabilities and production requirements. Next, a digital twin model is created for the pre-set coating station. By collecting the mechanical structure parameters, motion characteristics, control logic, and process flow of the station equipment, a twin coating station is constructed. This twin coating station can simulate the execution process of the actual coating process with high precision in a virtual environment, including film conveying, cylinder action, and changes in film application status. Subsequently, a set of initial coating control parameters (first coating control parameters) is randomly selected from the coating optimization space, including membrane tension, cylinder pressure, membrane conveying speed, and coating path parameters, and input into the twin coating station for simulation operation. During the simulation, the system records the dynamic bonding state of the membrane during coating, the contact between the membrane and the battery surface, and the overall operation time, generating the first coating simulation result. Based on the first coating simulation result, the coating quality is analyzed and scored through a built-in quality evaluation model. The quality evaluation indicators include coating accuracy (deviation), coating integrity (whether the coverage is uniform without wrinkles or bubbles), resource utilization (the effect of tension and pressure parameters), and operation efficiency (coating completion time). After the analysis, a first quality index is generated and compared with preset quality constraints. If the first quality index meets the preset quality constraints (such as coating accuracy deviation being less than the allowable range, membrane bonding being complete and without obvious defects, and efficiency reaching production standards), then the first coating control parameter is determined as the target coating control parameter. Otherwise, the system will continue to select new coating control parameters in the coating optimization space, repeating the above simulation and quality evaluation process until control parameters that meet the preset quality constraints are found. Ultimately, the generated target coating control parameters will serve as the basis for the actual coating process, and will be used in actual operation through the control platform of the cylinder servo component to ensure that the battery coating process achieves high precision, high efficiency, and high quality requirements.

[0031] Furthermore, establishing a space for optimizing the coating includes:

[0032] Obtain a set of battery pack control parameter types; obtain a similar packing station to the preset packing station; using the set of battery pack control parameter types and the packing requirement parameters as constraints, connect the control platform of the similar packing station and collect several battery packing process control records; construct the packing optimization space using the several battery packing process control records.

[0033] Specifically, the coating requirements of the target battery are analyzed to identify the set of control parameters needed for optimization, including film conveying speed, film tension, cylinder pressure, cylinder displacement range, and film application path parameters. Next, similar coating stations are identified by examining equipment files and production line layouts to find stations with identical structures and control logic. Historical operational data from these stations provides rich reference information, expanding the diversity of the optimization space. Then, using the set of battery coating control parameters and coating requirements as constraints, the control platform of the similar coating stations is connected via a Manufacturing Execution System (MES) to retrieve historical coating process control records. Several historical coating process control records are collected from these stations, each containing specific coating input parameters (such as film tension, cylinder pressure, film application speed, and path) and corresponding output results (such as film application accuracy, coating integrity, and work efficiency). Simultaneously, the collected data is cleaned and verified to remove outliers and invalid data, ensuring accuracy and completeness. Finally, based on these selected historical process control records, a coating optimization space is constructed. This optimization space is presented in the form of a multi-dimensional parameter matrix, where each dimension corresponds to a control parameter type, and each data point in the matrix represents a set of historical coating process records.

[0034] Furthermore, based on the first membrane simulation results, a membrane quality evaluation is performed to generate a first quality index, including:

[0035] A coating quality evaluation index is constructed, wherein the coating quality evaluation index includes the adhesion tightness between the coating material and the battery surface, the smoothness of the coating surface, the integrity of the coating, the sealing performance of the coating, and the consistency of the coating thickness; the quality evaluation index values ​​corresponding to the coating quality evaluation index are extracted based on the first coating simulation results to generate the first quality index.

[0036] Specifically, based on the quality requirements of the coating process, a comprehensive coating quality evaluation index is constructed. This index includes the following: the tightness of the adhesion between the coating material and the battery surface (measuring the tightness of contact between the coating material and the battery surface to avoid gaps or air bubbles); the smoothness of the coating surface (evaluating the flatness of the coating surface to avoid wrinkles or ripples); the integrity of the coating (checking whether the coating completely covers the target area of ​​the battery to avoid exposed or uncovered areas); the sealing performance of the coating (evaluating the sealing effect of the coating to ensure the battery is protected from external environmental influences); and the uniformity of the coating thickness (detecting whether the thickness of the coating on the battery surface is uniform to avoid uneven thickness affecting battery performance or appearance). The actual values ​​of each quality evaluation indicator are extracted from the membrane simulation results: image processing algorithms are used to analyze the membrane bonding state and calculate the bonding tightness; the height difference of the simulated image on the membrane surface is analyzed to extract smoothness data; the actual area covered by the membrane in the simulated image and the target area are checked to calculate the membrane integrity; an airtightness test is simulated, and the membrane sealing performance is calculated through leakage data; the thickness distribution of the membrane material in the simulation data is extracted, and the standard deviation is calculated to evaluate the thickness consistency; the values ​​of the above five indicators are normalized so that they all range between 0 and 1, and weights are assigned according to process requirements; the normalized indicator values ​​are multiplied by the weights and then summed to generate the total score of the first quality indicator.

[0037] Furthermore, it also includes:

[0038] If the first quality indicator does not meet the preset quality constraints, the first coating control parameter is marked as taboo.

[0039] In the coating optimization space, a second coating control parameter without a taboo mark is randomly selected and input into the twin coating station for simulation to generate the second coating simulation result;

[0040] Based on the second coating simulation results, the coating quality is evaluated, and a second quality index is generated.

[0041] If the second quality index meets the preset quality constraints, the second coating control parameter is used as the target coating control parameter.

[0042] When the system evaluates the first quality indicator and finds that it does not meet the preset quality constraints (e.g., a score below 90), it marks the current first coating control parameters as taboos in the coating optimization space. The taboo marking operation is as follows: in the parameter matrix of the optimization space, the index of the set of coating control parameters is marked as disabled. This is achieved by setting a Boolean value (e.g., setting the disable mark to True) or adding a taboo weight (e.g., setting the weight to an extremely high value to reduce the selection probability), ensuring that this parameter combination will not be repeatedly selected in subsequent optimization processes. Subsequently, the system filters the set of coating control parameters that have not been marked as taboos from the coating optimization space and randomly selects a new set of parameters as the second coating control parameters (e.g., randomly selecting different combinations of membrane tension, cylinder pressure, and film application path parameters). Then, the second coating control parameters are input into the twin coating station for simulation. The twin-coating station performs simulated operations based on the new parameter combination, including membrane material delivery, film application path execution, and pressure adjustment, generating a second coating simulation result. This second simulation result outputs key data through the simulation model, such as membrane material adhesion tightness, surface smoothness, coating integrity, sealing performance, and thickness consistency. Based on the second coating simulation result, the results are analyzed according to the quality evaluation process, extracting the value of each coating quality evaluation index and calculating the total score of the second quality index using preset weights. If the second quality index meets the preset quality constraints (e.g., total score ≥ 90 points), the second coating control parameter is determined as the target coating control parameter and recorded in the control system for subsequent operations. If the second quality index still does not meet the preset standard, the second coating control parameter is marked as taboo, and the above process is repeated. The coating optimization space continuously selects coating control parameters that are not marked as taboo, performing simulation, quality evaluation, and taboo marking operations sequentially until the target coating control parameter that meets the quality requirements is found.

[0043] The target coating control parameters are input into the servo cylinder control platform of the cylinder servo assembly for control, and the film bonding status image data is collected in real time by a vision inspection device.

[0044] Through the system interface, the generated target coating control parameters (such as membrane conveying speed, tension, cylinder pressure, displacement range, and coating path) are input to the servo cylinder control platform. After parsing the parameters, the servo cylinder control platform adjusts the servo cylinder according to the set values. The cylinder servo component starts working under the command of the control platform. Through its high-precision servo control function, it ensures that the membrane is uniformly and tightly bonded to the battery surface, avoiding membrane looseness, wrinkles, or excessive tightness. During the coating process, a vision inspection device (such as an industrial camera or multispectral camera) installed at the workstation collects real-time image data of the membrane bonding status.

[0045] The image data of the membrane bonding state is input into a pre-trained membrane bonding detection model for analysis, and the bonding anomaly detection result is output.

[0046] The image data of membrane bonding status acquired by the vision inspection device is preprocessed to ensure that it meets the input requirements of the pre-trained detection model. The preprocessed image data is then input into the pre-trained membrane bonding detection model, which is based on machine learning or deep learning algorithms and has learned the characteristics of membrane bonding status through a large amount of training data, including bonding tightness, surface smoothness, and edge coverage. The detection model analyzes the input image data, extracts key features of the bonding status, and calculates relevant indicators, such as bonding deviation value (the distance between the actual bonding position of the membrane and the target position), surface smoothness score (smoothness score calculated based on the surface height difference distribution), coverage (the ratio of the detected membrane coverage area to the target area), and defect location annotation (locating the location of wrinkles, bubbles, or unbonded edges in the image). Based on the feature data extracted by the model analysis, the bonding anomaly detection results are output.

[0047] Furthermore, the membrane bonding state image data is input into a pre-trained membrane bonding detection model for analysis, and the bonding anomaly detection results are output, including:

[0048] Collect images of bonding anomalies; train the membrane bonding detection model with the images of bonding anomalies using a convolutional neural network until convergence; input the membrane bonding state image data into the converged membrane bonding detection model for analysis, and output the bonding anomaly detection result.

[0049] In the actual film coating process, image samples of various film coating states are first collected using a visual inspection device, including normal and abnormal bonding samples. Abnormal samples must cover common defect types, such as uncovered edges, film wrinkles, and bubbles or gaps. Simultaneously, the collected image samples are labeled to clarify the type, location, and severity of the anomalies. Subsequently, all samples undergo standardized preprocessing, including resolution adjustment, noise reduction, and normalization, unifying the sample data into a standard format acceptable to the model. Based on this, a film bonding detection model is constructed using a convolutional neural network (CNN). The model includes multiple convolutional layers for extracting edge features, local bonding features, and texture features; pooling layers for dimensionality reduction to retain key information; and fully connected layers for outputting classification results. The preprocessed samples are divided into training, validation, and test sets. The model is trained using a cross-entropy loss function and the Adam optimizer. Parameters are iteratively optimized by inputting batches of training data until the accuracy and loss value on the validation set reach the convergence standard (e.g., accuracy above 95%). Finally, the converged model is saved for real-time detection. In the actual film coating operation, the real-time acquired images of the film's bonding status are preprocessed and then input into the trained detection model. After analyzing the image features, the model outputs detection results, including anomaly type (such as wrinkles or bubbles), anomaly location (such as rectangular box annotations), and anomaly score (based on region area and severity).

[0050] The control parameters of the target coating are adjusted based on the detection results of the adhesion anomaly.

[0051] After the membrane bonding detection model outputs bonding anomaly detection results, the system first analyzes the results, including key information such as anomaly type, anomaly location, and anomaly score. For example, the anomaly type might be membrane wrinkles, edge misalignment, or bubbles; the anomaly location is marked with coordinates to indicate the specific area; and the anomaly score quantifies the severity of the anomaly. After analysis, the system generates corresponding control and adjustment strategies based on the anomaly type and location, combined with the preset adjustment rules of the bonding process. Specifically, if wrinkles are detected in the bonding area, the system will improve the membrane bonding tightness by increasing or decreasing cylinder pressure, while adjusting the membrane tension parameters to alleviate the wrinkles; if edge misalignment is detected, the system will modify the cylinder displacement parameters or the bonding path to recalibrate the membrane position; if bubbles are detected, the system will increase local pressure or extend the bonding time to make the membrane adhere more tightly to the battery surface. The system determines the adjustment range based on the anomaly score; the higher the score, the larger the adjustment range, ensuring rapid correction of the anomaly.

[0052] In summary, the embodiments of this application have at least the following technical effects:

[0053] When the target battery is conveyed to the preset coating station, its coating requirement parameters are received. The preset coating station controls the film feeding and coating process via a cylinder servo assembly. Next, the coating control parameters are optimized based on the coating requirement parameters to generate target coating control parameters. These target coating control parameters are then input into the servo cylinder control platform of the cylinder servo assembly for control, and real-time image data of the film bonding status is acquired via a vision inspection device. Furthermore, the image data of the film bonding status is input into a pre-trained film bonding detection model for analysis, outputting bonding anomaly detection results. Finally, the target coating control parameters are adjusted based on the bonding anomaly detection results. This solves the technical problems of unstable film bonding quality and low efficiency in the battery coating process of existing technologies, achieving the technical effect of improving coating quality stability and production efficiency.

[0054] Example 2, based on the same inventive concept as the high-precision battery packing operation method based on the cylinder servo component in the foregoing examples, such as... Figure 2 As shown, this application provides a high-precision battery pack coating system based on a cylinder servo assembly, wherein the system includes:

[0055] Data receiving module 11: When the target battery is delivered to the preset coating station, it receives the coating requirement parameters of the target battery, wherein the preset coating station controls the film delivery and coating through a cylinder servo component; Optimization module 12: Based on the coating requirement parameters, it optimizes the coating control parameters to generate target coating control parameters; Control module 13: It inputs the target coating control parameters into the servo cylinder control platform of the cylinder servo component for control, and collects the film bonding state image data in real time through a vision inspection device; Analysis module 14: It inputs the film bonding state image data into a pre-trained film bonding detection model for analysis and outputs the bonding anomaly detection result; Feedback adjustment module 15: Based on the bonding anomaly detection result, it adjusts the target coating control parameters.

[0056] Furthermore, the data receiving module 11 is used to perform the following method:

[0057] The preset wrapping station includes a U-shaped wrapping station and a circular wrapping station; wherein, both the U-shaped wrapping station and the circular wrapping station are transported in an equidistant transport mode, and the U-shaped wrapping station is located before the circular wrapping station.

[0058] Furthermore, the optimization module 12 is used to perform the following method:

[0059] Establish a coating optimization space; perform digital twin modeling on the preset coating station to establish a twin coating station; randomly select a first coating control parameter in the coating optimization space and input it into the twin coating station for simulation to generate a first coating simulation result; evaluate the coating quality based on the first coating simulation result to generate a first quality index; if the first quality index meets the preset quality constraints, use the first coating control parameter as the target coating control parameter.

[0060] Furthermore, the optimization module 12 is used to perform the following method:

[0061] Obtain a set of battery pack control parameter types; obtain a similar packing station to the preset packing station; using the set of battery pack control parameter types and the packing requirement parameters as constraints, connect the control platform of the similar packing station and collect several battery packing process control records; construct the packing optimization space using the several battery packing process control records.

[0062] Furthermore, the optimization module 12 is used to perform the following method:

[0063] A coating quality evaluation index is constructed, wherein the coating quality evaluation index includes the adhesion tightness between the coating material and the battery surface, the smoothness of the coating surface, the integrity of the coating, the sealing performance of the coating, and the consistency of the coating thickness; the quality evaluation index values ​​corresponding to the coating quality evaluation index are extracted based on the first coating simulation results to generate the first quality index.

[0064] Furthermore, the optimization module 12 is used to perform the following method:

[0065] If the first quality indicator does not meet the preset quality constraints, the first coating control parameter is marked as taboo; a second coating control parameter without taboo marking is randomly selected in the coating optimization space and input into the twin coating station for simulation to generate a second coating simulation result; the coating quality is evaluated based on the second coating simulation result to generate a second quality indicator; if the second quality indicator meets the preset quality constraints, the second coating control parameter is used as the target coating control parameter.

[0066] Furthermore, the analysis module 14 is used to perform the following methods:

[0067] Collect images of bonding anomalies; train the membrane bonding detection model with the images of bonding anomalies using a convolutional neural network until convergence; input the membrane bonding state image data into the converged membrane bonding detection model for analysis, and output the bonding anomaly detection result.

[0068] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0069] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0070] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A high-precision battery pack coating operation method based on a cylinder servo assembly, characterized in that, The method includes: When the target battery is delivered to the preset coating station, the coating requirement parameters of the target battery are received. The preset coating station controls the film delivery and coating through a cylinder servo component. Based on the coating requirement parameters, the coating control parameters are optimized to generate target coating control parameters; The target coating control parameters are input into the servo cylinder control platform of the cylinder servo assembly for control, and the image data of the film bonding status are collected in real time by the vision inspection device. The image data of the membrane bonding state is input into a pre-trained membrane bonding detection model for analysis, and the bonding anomaly detection result is output. The target coating control parameters are adjusted based on the detection results of the adhesion anomaly. Based on the coating requirement parameters, the coating control parameters are optimized to generate target coating control parameters, including: Establish a space for optimizing the coating; Digital twin modeling is performed on the preset coating station to establish a twin coating station; The first coating control parameter is randomly selected in the coating optimization space and input into the twin coating station for simulation to generate the first coating simulation result; Based on the first membrane simulation results, the membrane quality is evaluated, and a first quality index is generated. If the first quality indicator meets the preset quality constraints, the first coating control parameter is used as the target coating control parameter.

2. The high-precision battery pack coating method based on a cylinder servo assembly as described in claim 1, characterized in that, The preset wrapping station includes a U-shaped wrapping station and a U-shaped wrapping station; Both the U-shaped package station and the loop-shaped package station are transported using an equidistant transport method, and the U-shaped package station is located before the loop-shaped package station.

3. The high-precision battery pack coating method based on a cylinder servo assembly as described in claim 1, characterized in that, Establish a space for optimizing the envelope, including: Obtain the set of battery pack membrane control parameter types; Obtain a similar coating station to the preset coating station; Using the set of battery pack control parameter types and the packing requirement parameters as constraints, the control platform of the same type packing station is connected to collect several battery packing process control records. The aforementioned battery packing process control records are used to construct the packing optimization space.

4. The high-precision battery pack coating method based on a cylinder servo assembly as described in claim 1, characterized in that, Based on the first membrane simulation results, the membrane quality is evaluated, and a first quality index is generated, including: A coating quality evaluation index is constructed, wherein the coating quality evaluation index includes the tightness of the adhesion between the coating material and the battery surface, the smoothness of the coating surface, the integrity of the coating, the sealing of the coating, and the consistency of the coating thickness. Based on the first membrane simulation results, extract the quality evaluation index values ​​corresponding to the membrane quality evaluation index to generate the first quality index.

5. The high-precision battery pack coating method based on a cylinder servo assembly as described in claim 4, characterized in that, Also includes: If the first quality indicator does not meet the preset quality constraints, the first coating control parameter is marked as taboo. In the coating optimization space, a second coating control parameter without a taboo mark is randomly selected and input into the twin coating station for simulation to generate the second coating simulation result; Based on the second coating simulation results, the coating quality is evaluated, and a second quality index is generated. If the second quality index meets the preset quality constraints, the second coating control parameter is used as the target coating control parameter.

6. The high-precision battery pack coating method based on a cylinder servo assembly as described in claim 1, characterized in that, The membrane bonding state image data is input into a pre-trained membrane bonding detection model for analysis, and the bonding anomaly detection results are output, including: Collect images with abnormal fit; The membrane material bonding detection model is trained until convergence using the bonding anomaly image samples based on a convolutional neural network. The membrane bonding state image data is input into a converged membrane bonding detection model for analysis, and the bonding anomaly detection result is output.

7. A high-precision battery pack coating system based on a cylinder servo assembly, characterized in that, The system is used to implement the high-precision battery packing operation method based on a cylinder servo assembly as described in any one of claims 1-6, the system comprising: Data receiving module: When the target battery is delivered to the preset coating station, it receives the coating requirement parameters of the target battery, wherein the preset coating station controls the film delivery and coating through a cylinder servo component; Optimization module: Based on the coating requirement parameters, optimize the coating control parameters and generate target coating control parameters; Control module: Inputs the target coating control parameters into the servo cylinder control platform of the cylinder servo assembly for control, and collects real-time image data of the film bonding status through a vision inspection device; Analysis module: Inputs the membrane bonding state image data into the pre-trained membrane bonding detection model for analysis and outputs bonding anomaly detection results; Feedback adjustment module: Adjusts the target coating control parameters based on the adhesion anomaly detection results.

Images