Battery cell size measuring method and system based on machine vision

Abstract

The invention discloses a machine vision-based battery cell size measurement method and system, and relates to the technical field of battery cell size measurement, and the method comprises the steps: calling a 3D visual module to capture the 3D point cloud of a battery cell, and constructing an initial point cloud set; battery cell deformation data under typical working conditions are obtained, deformation analysis is carried out, and a clustering deformation field is constructed; arranging and optimizing strain gauges, configuring the strain gauges according to an arranging and optimizing result, executing battery cell monitoring, and generating a monitoring deformation data set; and after the initial point cloud set is registered, contour position coordinates are established, coordinate compensation is carried out based on the monitored deformation data set, and a cell size measurement result is generated according to a coordinate compensation result. The technical problems that in the prior art, the dynamic deformation characteristic of the battery cell under the real charging and discharging cycle working condition is difficult to capture, so that the data fusion precision and real-time performance are insufficient, and the size measurement precision is poor are solved, and the technical effect of improving the battery cell size measurement precision and the battery management safety is achieved.

Description

Technical Field

[0001] This application relates to the technical field of cell size measurement, and specifically relates to a cell size measurement method and system based on machine vision. Background Art

[0002] As the basic unit of a battery, the accurate measurement of the geometric size of a cell directly affects the assembly accuracy, thermal management efficiency, and long-term cycle stability of the battery module. The measurement results of traditional cell sizes are extremely vulnerable to interference from other factors, with large accuracy deviations, making it difficult to meet the current high-precision requirements for cell manufacturing. Even if some contact measurement devices are introduced, although the accuracy is improved to some extent, the data fusion accuracy is insufficient, and potential damages such as scratches and abrasions will be caused to the cell surface due to contact measurement, affecting the performance and life of the cell; in addition, under dynamic working conditions (such as charging and discharging, temperature changes, mechanical vibrations), the cell is prone to local deformation, resulting in the accumulation of size deviations, making it difficult to capture dynamic deformation characteristics, unable to accurately characterize the three-dimensional deformation distribution and complex surface characteristics of the cell, and even causing safety hazards such as thermal runaway.

[0003] Therefore, in the current related technologies, there are technical problems that it is difficult to capture the dynamic deformation characteristics of the cell under real charge and discharge cycle conditions, which in turn leads to insufficient data fusion accuracy and real-time performance, and poor size measurement accuracy. Summary of the Invention

[0004] This application provides a cell size measurement method and system based on machine vision, which solves the technical problems in the prior art that it is difficult to capture the dynamic deformation characteristics of the cell under real charge and discharge cycle conditions, which in turn leads to insufficient data fusion accuracy and real-time performance, and poor size measurement accuracy, and achieves the technical effect of improving the accuracy of cell size measurement and the safety of battery management.

[0005] This application provides a cell size measurement method based on machine vision. The method includes: calling a 3D vision module to capture the 3D point cloud of the cell and constructing an initial point cloud set; establishing typical working conditions of the cell, obtaining cell deformation data under the typical working conditions, performing deformation analysis on the cell deformation data, and constructing a clustering deformation field, where the clustering deformation field includes a high-strain region, a transition region, and a low-strain region; performing optimization of the layout of strain gauges in the clustering deformation field, establishing an optimization result of the layout, and configuring the strain gauges with the optimization result of the layout, performing cell monitoring to generate a monitored deformation data set, and the objective function of the layout optimization includes a monitoring accuracy function, a cost constraint function, and a safety constraint function; after registering the initial point cloud set, performing contour recognition to establish contour position coordinates, performing coordinate compensation on the contour position coordinates based on the monitored deformation data set, and generating a cell size measurement result according to the coordinate compensation result.

[0006] In a possible implementation manner, the machine vision-based cell size measurement method further performs the following processing: establishing an objective function for layout optimization, and performing layout optimization according to the objective function. The objective function is as follows: ; Wherein, represents the objective function, represents the strain gauge set, represents the strain gauge set the number of strain gauges inside, represents based on the strain gauge set the full-field deformation reconstruction error under the scheme, represents the maximum allowable reconstruction error, the upper limit of the number of strain gauges, represents the safety constraint function, , , are the weight coefficients of the monitoring precision term, cost constraint term, and safety constraint term respectively.

[0007] In a possible implementation manner, the machine vision-based cell size measurement method further performs the following processing: The monitoring precision function is as follows: ; Wherein, is the total number of discrete grid points on the cell surface, represents any grid point, represents the th grid point's true deformation amount, represents the deformation amount inverted from the strain data at the position of the strain gauge set .

[0008] In a possible implementation manner, the machine vision-based cell size measurement method further performs the following processing: configuring the initial distribution of strain gauges through historical experience data, establishing an initial distribution set, and each subset in the initial distribution set represents a distribution scheme of a set of strain gauges; performing fitness analysis of strain gauges under the initial distribution set according to the objective function to generate a fitness analysis result; performing subset region fitness comparison within the initial distribution set according to the regional division of the clustering deformation field, and generating a regional optimization scheme according to the region fitness comparison result and the fitness analysis result; performing iterative update of the initial distribution set according to the regional optimization scheme to complete the layout optimization of strain gauges.

[0009] In a possible implementation manner, the machine vision-based battery cell size measurement method further performs the following processing: establishing an optimized reference target value according to the fitness analysis result; performing trust superposition based on the regional fitness comparison result and the reference target value, and determining an optimization target according to the trust superposition result; performing update analysis on the initial distribution set with the optimization target, and establishing a regional optimization scheme.

[0010] In a possible implementation manner, the machine vision-based battery cell size measurement method further performs the following processing: activating a laser sensor, performing laser data acquisition of the battery cell, and establishing an additional data set; after extracting the key point position coordinates according to the additional data set, performing data registration on the data of the initial point cloud based on the key point position coordinates, and performing contour recognition according to the data registration result to establish contour position coordinates.

[0011] In a possible implementation manner, the machine vision-based battery cell size measurement method further performs the following processing: the 3D vision module and the laser sensor are installed at the production line station, and the strain gauge is integrated into the battery cell carrier.

[0012] In a possible implementation manner, the machine vision-based battery cell size measurement method further performs the following processing: configuring a preset cycle range, and when a time node triggers the preset cycle range, performing cycle adaptive calibration, where the cycle adaptive calibration includes performing a loading test using a standard battery cell, generating a loading test result, calculating a drift compensation coefficient according to the loading test result, and performing measurement calibration management according to the drift compensation coefficient.

[0013] In a possible implementation manner, the machine vision-based battery cell size measurement method further performs the following processing: performing size anomaly recognition with the battery cell size measurement result, and establishing a size anomaly recognition warning; performing anomaly reporting management according to the size anomaly recognition warning.

[0014] The present application also provides a battery cell size measurement system based on machine vision, including: an initial point cloud set construction module, which is used to call a 3D vision module to capture the 3D point cloud of the battery cell and construct an initial point cloud set; a clustering deformation field construction module, which is used to establish typical working conditions of the battery cell, obtain the battery cell deformation data under the typical working conditions, perform deformation analysis on the battery cell deformation data, and construct a clustering deformation field, where the clustering deformation field includes a high strain region, a transition region, and a low strain region; a monitoring deformation data set generation module, which is used to optimize the layout of strain gauges in the clustering deformation field, establish the layout optimization result, configure the strain gauges with the layout optimization result, perform battery cell monitoring, generate a monitoring deformation data set, and the objective function of the layout optimization includes a monitoring accuracy function, a cost constraint function, and a safety constraint function; a battery cell size measurement result generation module, which is used to perform registration on the initial point cloud set, then perform contour recognition, establish contour position coordinates, perform coordinate compensation on the contour position coordinates based on the monitoring deformation data set, and generate a battery cell size measurement result according to the coordinate compensation result.

[0015] It is intended to call a 3D vision module to capture the 3D point cloud of the battery cell through the battery cell size measurement method and system based on machine vision proposed in the present application, construct an initial point cloud set; obtain the battery cell deformation data under typical working conditions, perform deformation analysis, and construct a clustering deformation field; perform optimization on the layout of strain gauges, configure the strain gauges with the layout optimization result, perform battery cell monitoring, and generate a monitoring deformation data set; after registering the initial point cloud set, establish contour position coordinates, perform coordinate compensation based on the monitoring deformation data set, and generate a battery cell size measurement result according to the coordinate compensation result. This solves the technical problems in the prior art that it is difficult to capture the dynamic deformation characteristics of the battery cell under real charge and discharge cycle working conditions, which in turn leads to insufficient data fusion accuracy and real-time performance and poor size measurement accuracy, and achieves the technical effect of improving the accuracy of battery cell size measurement and the safety of battery management. Description of the Drawings

[0016] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic flowchart of the battery cell size measurement method based on machine vision provided by the embodiments of the present application.

[0018] Figure 2 It is a schematic structural diagram of the battery cell size measurement system based on machine vision provided by the embodiments of the present application.

[0019] Description of reference numerals: Initial point cloud set construction module 10, clustering deformation field construction module 20, monitoring deformation data set generation module 30, battery cell size measurement result generation module 40. Detailed implementation manners

[0020] The above description is only an overview of the technical solution of the present application. In order to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0022] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0023] The embodiment of the present application provides a method for measuring the size of a battery cell based on machine vision, as Figure 1 shown, the method includes: Step S100, call a 3D vision module to capture the 3D point cloud of the battery cell and construct an initial point cloud set.

[0024] Preferably, the 3D vision module is a device capable of acquiring three-dimensional spatial information of an object. It usually integrates multiple sensors, including structured light 3D cameras, lidars, etc. Different types of 3D vision modules have different working principles, but the purpose is to obtain the three-dimensional coordinate information of the object surface. Among them, the structured light 3D camera projects a specific structured light pattern (such as stripes, coded patterns, etc.) onto the object surface, and then uses the camera to capture the pattern modulated by the object surface. According to the deformation of the pattern, the three-dimensional coordinates of each point on the object surface are calculated using the triangulation principle; the lidar calculates the distance between each point on the object surface and the sensor by emitting a laser beam and measuring the time (time of flight, TOF) from the emission of the laser to its reflection back to the receiver. Combining the emission angle information of the laser beam, the three-dimensional coordinates of each point on the object surface can be obtained.

[0025] Preferably, align the 3D vision module with the battery cell to be measured and start the module for data acquisition. During the acquisition process, the 3D vision module will emit light (such as laser or structured light) onto the surface of the battery cell and receive the light reflected back from the surface of the battery cell. By analyzing and processing the reflected light, the module can calculate the coordinates of each point on the surface of the battery cell in three-dimensional space. These coordinate points form the 3D point cloud data of the battery cell, that is, a set of a large number of discrete three-dimensional coordinate points. Each point represents a specific position on the surface of the battery cell, reflecting the shape and contour information of the surface of the battery cell; after obtaining the 3D point cloud data of the battery cell, organize these data to form an initial point cloud set, which may include screening the collected original point cloud data to remove those that are obviously incorrect or do not meet the requirements; store the screened point cloud data in a certain format. Common point cloud data storage formats include PLY, XYZ, OBJ, etc.; group, sort the data according to certain rules, or establish an index structure to improve the access efficiency of the data; finally, realize calling the 3D vision module to capture the 3D point cloud of the battery cell and construct an initial point cloud set, providing a data basis for subsequent battery cell size measurement and analysis.

[0026] Step S200, establish a typical working condition of the battery cell, obtain the deformation data of the battery cell under the typical working condition, perform deformation analysis on the deformation data of the battery cell, and construct a clustering deformation field, where the clustering deformation field includes a high-strain region, a transition region, and a low-strain region.

[0027] Preferably, establish the typical working conditions of the battery cell, such as different charge and discharge states, ambient temperatures, usage frequencies, load sizes, etc. Specifically, consider different charge and discharge rates, depths, etc. For example, the states of the battery cell are different during fast charging and slow charging, and deep charge and discharge and shallow charge and discharge will also make the battery cell exhibit different characteristics. Taking the power battery of an electric vehicle as an example, when accelerating rapidly, the battery discharges with a large current, while in urban congested road conditions, it may discharge with a small current or be in a charging state, all of which belong to different charge and discharge working conditions; the ambient temperature has a significant impact on the battery cell. High temperature will accelerate the chemical reaction rate but may cause performance degradation, while low temperature may cause the battery capacity to decrease and the internal resistance to increase. For example, the battery cells of electronic devices used in desert areas have to face high-temperature and dry environments; for devices used in polar regions, the battery cells have to withstand low-temperature environments; it is determined according to the application scenario of the battery cell. For example, the working conditions of a mobile phone battery are different when frequently using high-energy-consuming applications (such as games) and when only making simple calls and sending text messages; the working modes of energy storage batteries also vary in the scenarios of power grid peak shaving and household energy storage; by comprehensively analyzing these factors, several combinations of working conditions are determined to form the typical working conditions of the battery cell.

[0028] Preferably, under different typical working conditions, complex physical and chemical changes will occur inside the battery cell, resulting in changes in the external shape and size of the battery cell, that is, deformation occurs. Obtain the deformation data of the battery cell under typical working conditions. For example, during large-current charge and discharge, the rapid insertion and extraction of lithium ions inside the battery cell may cause the electrode material to expand or contract; temperature changes will cause thermal expansion and contraction of materials, etc. Specifically, strain gauges, displacement sensors, etc. can be installed on the surface or inside of the battery cell. The strain gauge can convert the strain on the surface of the battery cell into an electrical signal, and the strain condition of the battery cell can be obtained by measuring the change in the electrical signal. The displacement sensor can directly measure the displacement change of a certain part of the battery cell to obtain the deformation data; using digital image correlation technology (DIC), by analyzing the images of the surface of the battery cell under different working conditions, the displacement and strain of each point on the surface can be calculated; using a 3D vision module, the 3D point cloud data of the battery cell can be captured again under different working conditions, and by comparing the initial point cloud set, the change in the point cloud data can be analyzed to obtain the deformation information of the battery cell, including overall expansion, contraction, and local deformation, etc.

[0029] Preferably, deformation analysis is performed on the cell deformation data to construct a clustering deformation field, and high-strain regions, transition regions, and low-strain regions are divided to deeply understand the deformation characteristics of the cell under different working conditions. Specifically, the obtained cell deformation data is preprocessed, such as removing noise and filling missing values, to improve the data quality. Then, features related to deformation, such as the magnitude, change rate, and direction of strain, are extracted from these data to more intuitively reflect the deformation of the cell. Then, clustering analysis is performed, that is, different regions of the cell are divided into different categories based on the similarity of the data. Taking each position on the cell surface or inside as a sample point and the extracted deformation features as variables, a clustering algorithm is used to perform clustering operations on all sample points. Through continuous iteration and optimization, the entire region of the cell is finally divided into different clustering regions, and each region corresponds to a specific deformation feature. For example, in the K-Means algorithm, according to the set number of clusters (assuming 3 clusters, corresponding to the high-strain region, transition region, and low-strain region), the deformation data points are assigned to different clusters, so that the data points within the same cluster have high similarity, while the data points between different clusters are quite different. Among them, the high-strain region refers to the region where the cell undergoes large deformation under specific working conditions. The performance and stability of the cell may be greatly affected, and problems such as fatigue and cracks are more likely to occur. The low-strain region is opposite to the high-strain region. The deformation amount of this region is relatively small under various typical working conditions because the mechanical properties of the material are good or the stress borne inside the cell is small. Therefore, it can maintain a relatively stable shape when the working conditions change, and the impact on the overall performance of the cell is relatively small. The transition region is between the high-strain region and the low-strain region, and its strain degree is between the two, playing a connecting and transitional role. The strain change in the transition region is relatively complex, being affected by both the high-strain region and interacting with the low-strain region. Its strain characteristics may fluctuate within a certain range as the working conditions change.

[0030] Step S300, optimize the layout of strain gauges in the clustering deformation field, establish the optimization result of the layout, and configure the strain gauges with the optimization result of the layout, perform cell monitoring, and generate a monitored deformation data set. The objective function of the layout optimization includes a monitoring accuracy function, a cost constraint function, and a safety constraint function.

[0031] Preferably, a strain gauge is a sensor used to measure the surface strain of an object. By reasonably arranging strain gauges on the surface of the battery cell, the deformation of the battery cell under different working conditions can be monitored in real time. The clustering deformation field has divided the battery cell into a high-strain region, a transition region, and a low-strain region. The deformation characteristics of different regions are different. Determine the optimal arrangement position of the strain gauges according to the deformation characteristics, that is, continuously try different strain gauge arrangement schemes in different regions of the clustering deformation field, and finally find the optimal arrangement method by evaluating the advantages and disadvantages of each scheme. Specifically, according to the preset objective functions (i.e., the monitoring accuracy function, the cost constraint function, and the safety constraint function), evaluate each strain gauge arrangement scheme, and judge whether the scheme meets the requirements by calculating the values of different schemes under the objective functions. Finally, select the scheme that makes the objective function reach the optimal value as the final arrangement optimization result to clarify the specific position and quantity of the strain gauges in the clustering deformation field.

[0032] Preferably, according to the arrangement optimization result, accurately install the strain gauges at the specified positions on the surface of the battery cell to ensure good contact between the strain gauges and the surface of the battery cell, so as to ensure accurate measurement of the strain of the battery cell; after completing the strain gauge configuration, start real-time monitoring of the battery cell, convert the strain on the surface of the battery cell into an electrical signal and process and analyze it, so as to obtain the deformation information of the battery cell under different working conditions, continuously collect the electrical signals output by the strain gauges, and convert them into corresponding strain values to generate a monitoring deformation data set, which may include corresponding working condition information (such as charge and discharge state, ambient temperature, etc.). Among them, the monitoring accuracy function is used to measure the accuracy of the strain gauge arrangement scheme for monitoring the deformation of the battery cell, and is usually related to the position and quantity of the strain gauges and the characteristics of the clustering deformation field. For example, reasonably increasing the quantity of strain gauges in the high-strain region can improve the monitoring accuracy of the deformation in this region. By optimizing the monitoring accuracy function, the strain gauges can capture the deformation information of the battery cell more accurately; the cost constraint function is used to control the cost of the entire monitoring system. On the premise of meeting the monitoring accuracy requirements, try to reduce the use quantity of strain gauges or select strain gauge products with high cost performance. For example, the arrangement of strain gauges can be appropriately reduced in the low-strain region to reduce costs; the safety constraint function is used to ensure that the arrangement of strain gauges will not have a negative impact on the safety of the battery cell. For example, avoid arranging strain gauges too densely in the key parts or areas prone to failure of the battery cell, so as not to affect the normal operation of the battery cell or increase safety risks such as short circuits. By comprehensively considering the three objective functions, optimize the arrangement of strain gauges to achieve efficient, accurate and safe monitoring of the battery cell.

[0033] Furthermore, step S300 further includes establishing an objective function for arrangement optimization, and performing arrangement optimization according to the objective function. The objective function is as follows: ; Among them, Characterize the objective function, Characterize the strain gauge set, Characterize the strain gauge set The number of strain gauges inside, Characterize based on the strain gauge set The full-field deformation reconstruction error under the scheme, Characterize the maximum allowable reconstruction error, The upper limit of the number of strain gauges, Characterize the safety constraint function, 、 、 Are the weight coefficients of the monitoring precision term, cost constraint term, and safety constraint term respectively.

[0034] Preferably, Measure the monitoring accuracy, reflect the ratio of the full-field deformation reconstruction error to the maximum allowable reconstruction error under the scheme based on the strain gauge set S. The smaller this value, the higher the monitoring accuracy; Measure the cost constraint, reflect the ratio of the number of strain gauges in the strain gauge set S to the upper limit of the number, and be used to control the cost; Characterize the safety constraint function, and be used to evaluate the impact of the strain gauge layout on the safety of the battery cell; Be used to calculate the full-field deformation reconstruction error under the scheme based on the strain gauge set S. By calculating the true deformation of the discrete grid points on the surface of the battery cell And the deformation obtained by inverting the strain data at the positions of the strain gauge set S The average value of the sum of squares of the differences between them is used to measure. The smaller this value, the closer the inverted deformation is to the true value, and the higher the monitoring accuracy.

[0035] Furthermore, step S300 further includes that the monitoring accuracy function is as follows: ; Wherein, Is the total number of discrete grid points on the surface of the battery cell, Characterize any grid point, Characterize the th grid point's true deformation, Characterize the deformation obtained by inverting the strain data at the positions of the strain gauge set S.

[0036] Further, step S300 further includes step S310 of configuring the initial distribution of strain gauges through historical experience data to establish an initial distribution set, where each subset in the initial distribution set represents a distribution scheme of a set of strain gauges; step S320 of performing fitness analysis of the strain gauges under the initial distribution set according to the objective function to generate a fitness analysis result; step S330 of comparing the subset region fitness within the initial distribution set according to the regional division of the clustering deformation field, and generating a regional optimization scheme according to the regional fitness comparison result and the fitness analysis result; step S340 of performing iterative update of the initial distribution set according to the regional optimization scheme to complete the layout optimization of the strain gauges.

[0037] Preferably, the historical experience data includes information such as the deformation characteristics of the battery cells under different working conditions and the monitoring effects corresponding to different strain gauge distribution schemes. According to the historical experience data, a variety of different strain gauge distribution schemes are designed, and each scheme corresponds to a layout method of the strain gauges on the surface of the battery cell, and then combined to form an initial distribution set. Among them, each subset in the initial distribution set represents a distribution scheme of a set of strain gauges, including information such as the specific positions and quantities of the strain gauges; the fitness analysis calculation is performed on each subset (i.e., each set of strain gauge distribution schemes) in the initial distribution set by using the objective function, that is, according to the calculation result of the objective function, the fitness of the scheme is evaluated. The higher the fitness, the better the scheme meets our requirements and the better the comprehensive performance in terms of monitoring accuracy, cost, and safety. By analyzing the fitness of all schemes, a fitness analysis result is generated, usually including the fitness scores of each scheme and the corresponding rankings.

[0038] Preferably, for each subset in the initial distribution set, analyze the distribution of strain gauges in each region of the clustering deformation field, and evaluate the fitness of each scheme in different regions. For example, in the high-strain region, whether the distribution of strain gauges can accurately capture the large deformation in this region; in the low-strain region, whether the over-dense layout is avoided to reduce costs. Through the comparison of regional fitness, obtain the fitness evaluation of each scheme in different regions (i.e., the result of regional fitness comparison). Then, comprehensively consider the result of regional fitness comparison and the previous fitness analysis result, find the scheme that performs well in each region and overall, and generate a regional optimization scheme, which can improve the monitoring effect of strain gauges in different regions while taking into account cost and safety factors. Finally, apply the regional optimization scheme to the initial distribution set and update the subsets in it. For example, adjust the position or quantity of strain gauges in some schemes according to the optimization scheme, and repeat the above fitness analysis, regional fitness comparison, and iterative update process until certain stop conditions are met, such as the fitness score no longer improves significantly, reaching the preset number of iterations, etc. Finally, complete the optimization of the layout of strain gauges. The scheme with the highest fitness in the initial distribution set is the final layout scheme of strain gauges, thus realizing the efficient and accurate monitoring of the battery cell.

[0039] Further, step S330 further includes step S331 of establishing a benchmark target value for optimization according to the fitness analysis result; step S332 of performing trust superposition based on the regional fitness comparison result and the benchmark target value, and determining an optimization target according to the trust superposition result; step S333 of performing update analysis on the initial distribution set with the optimization target and establishing a regional optimization scheme.

[0040] Preferably, based on multiple fitness values in the fitness analysis result, set a reasonable threshold as the benchmark target value to measure whether a new scheme is better. Then, perform trust superposition on the regional fitness comparison result and the benchmark target value. For example, through a certain weight allocation method, perform weighted summation on the regional fitness and the benchmark target value, and give a higher weight to the scheme that performs well in the high-strain region because the accurate monitoring of the high-strain region is crucial for the overall monitoring effect. Through trust superposition, obtain a new value that takes into account the regional performance and overall fitness, and then determine a clear optimization target according to the trust superposition result, that is, a more comprehensive and accurate evaluation criterion that takes into account the fitness of the scheme in each region and combines the previously established benchmark target value.

[0041] Preferably, guided by the optimization objective, each subset in the initial distribution set is re-evaluated and analyzed to compare the gap between each solution and the optimization objective, and identify areas for improvement. For example, if the fitness of a certain solution is low in a certain area, or the overall fitness is significantly different from the optimization objective, then adjust the solution. Finally, based on the results of the updated analysis, optimize the solutions in the initial distribution set, including adjusting parameters such as the position and quantity of strain gauges according to the characteristics and requirements of different regions, to obtain a region-optimized solution that is closer to the optimization objective in each region and overall, thereby improving the overall performance of the strain gauge layout solution and ensuring that the final strain gauge distribution solution can achieve good monitoring effects in different regions and overall, while taking into account factors such as cost and safety.

[0042] Step S400, after registering the initial point cloud set, perform contour recognition to establish contour position coordinates, and perform coordinate compensation for the contour position coordinates based on the monitored deformation data set, and generate a cell size measurement result according to the coordinate compensation result.

[0043] Preferably, when obtaining the initial point cloud set, due to factors such as the measurement error of the 3D vision module and the difference in the placement position of the cell, the point cloud data may have a position deviation. Then register the initial point cloud set, that is, align the point cloud data obtained from different perspectives or at different times so that they have an accurate relative position relationship in the same coordinate system. For example, use the iterative closest point (ICP) algorithm for registration, and continuously iterate to find the optimal transformation matrix (including translation and rotation) between the two point clouds, so that the distance between the corresponding points of the two point clouds is minimized, thereby achieving accurate registration of the point cloud. Then perform contour recognition, that is, extract the outer edge information of the cell from the registered initial point cloud set. By recognizing the contour, clearly define the shape and range of the cell, that is, determine the boundary of the cell. For example, based on an edge detection algorithm such as the Canny edge detection algorithm, calculate the gradient value of each point in the point cloud data, and find the points with larger gradient values as edge points. Connecting these edge points forms the contour of the cell. Then establish accurate position coordinates for each point on the recognized contour. Usually, use the Cartesian coordinate system, with a certain fixed point as the origin, determine the directions of the X, Y, and Z coordinate axes, and convert the position information of each point on the contour into coordinate values in this coordinate system according to the registered point cloud data, so as to obtain a set of contour position coordinates, which accurately describes the position and shape of the cell contour in three-dimensional space.

[0044] Preferably, the battery cell will deform under various working conditions (such as charge and discharge, temperature change, etc.). The monitored deformation data set records the deformation information of the battery cell under different working conditions. Then, the contour position coordinates are corrected according to the monitored deformation data set, taking into account the actual deformation of the battery cell to obtain a more accurate measurement result of the battery cell size. Specifically, according to the deformation amounts (such as displacement, strain, etc.) of each point recorded in the monitored deformation data set, the contour position coordinates are adjusted accordingly. For example, if a certain point has a displacement ∆X in the X-axis direction under a certain working condition, then the X coordinate value of this point is added with ∆X to achieve coordinate compensation. Finally, various size parameters of the battery cell are calculated according to the compensated coordinate data. For example, by finding the coordinates of the corresponding end points on the contour and calculating the distance between them, the basic sizes such as the length, width, and height of the battery cell can be obtained, and then a battery cell size measurement result that accurately reflects the actual size of the battery cell is generated.

[0045] Further, step S400 further includes step S410 of activating the laser sensor, performing laser data acquisition of the battery cell, and establishing an additional data set; step S420 of, after extracting the key point position coordinates according to the additional data set, performing data registration on the initial point cloud data based on the key point position coordinates, and performing contour recognition according to the data registration result to establish the contour position coordinates.

[0046] Preferably, the laser sensor emits a laser beam and then receives the laser signal reflected from the surface of the battery cell. According to the time of flight (TOF) of the laser or the triangulation principle, the distance from the sensor to each point on the surface of the battery cell is calculated. By scanning the surface of the battery cell, a large amount of distance information can be obtained. The laser sensor is activated to scan the battery cell according to a predetermined path or mode, collecting the distance data of each point on the surface of the battery cell and the corresponding spatial position information, etc., to obtain an additional data set containing detailed geometric information on the surface of the battery cell. Analyze the data in the additional data set, and screen key points from multiple points reflecting the shape and structure characteristics of the battery cell, such as the vertices of the battery cell, edge turning points, points with large curvature changes, etc. After determining the key points, extract the position coordinates of these key points from the additional data set, including the X, Y, and Z coordinate values in the three-dimensional space, for subsequent data registration and contour recognition operations.

[0047] Preferably, data registration is performed on the initial point cloud set. That is, using the position coordinates of the key points as a reference, through a specific registration algorithm (such as the Iterative Closest Point algorithm ICP), the transformation relationship (including translation and rotation) between the initial point cloud set and the additional data set is calculated. Then, the initial point cloud set is subjected to the corresponding transformation to make it consistent with the additional data set in terms of spatial position. Finally, using image processing or point cloud processing algorithms, the outer contour of the battery cell is identified from the registered data. For example, by detecting the edge points in the point cloud data and connecting these edge points to form a closed curve, which is the contour of the battery cell, and then extracting the position coordinates of each point on the contour to form a contour position coordinate set.

[0048] Preferably, step S400 further includes that the 3D vision module and the laser sensor are installed at the production line station, and the strain gauge is integrated into the battery cell carrier.

[0049] Preferably, on the battery cell production line, a 3D vision module and a laser sensor are installed at a specific working position (production line station). The 3D vision module can obtain the three-dimensional spatial information of the battery cell surface and construct an initial point cloud set. The laser sensor can collect the laser data of the battery cell by emitting and receiving laser signals. Data collection of the battery cell can be carried out in a timely manner when the battery cell passes through this position during the production or detection process. The strain gauge is an element used to monitor the deformation of the battery cell and is integrally installed on the battery cell carrier. Among them, the battery cell carrier is a tool used to carry the battery cell during the production line transfer. Integrating the strain gauge onto it enables the strain gauge to real-time monitor the deformation data of the battery cell under various working conditions when the battery cell moves with the carrier and experiences different production links, and does not affect the normal transmission of the battery cell on the production line.

[0050] Further, step S400 further includes configuring a preset cycle range. When the time node triggers the preset cycle range, periodic adaptive calibration is performed. The periodic adaptive calibration includes using a standard battery cell for a loading test to generate a loading test result, calculating a drift compensation coefficient based on the loading test result, and performing measurement calibration management based on the drift compensation coefficient.

[0051] Preferably, a time cycle range is set according to production process requirements, equipment characteristics, detection accuracy requirements, etc. When the time advances to a time node that matches the preset cycle range, the system automatically starts the periodic adaptive calibration process, that is, according to the pre-set rules, calibration operations are started at specific time points to ensure that the equipment or measurement system always maintains the best working state and adapts to various changes and possible error accumulations in the production process. Among them, the periodic adaptive calibration includes using a standard battery cell for a loading test to generate a loading test result, calculating a drift compensation coefficient based on the loading test result, and performing measurement calibration management based on the drift compensation coefficient.

[0052] Preferably, a standard cell is a cell with precisely known characteristics and parameters, which serves as a benchmark to measure the performance of other cells and the accuracy of detection equipment. The standard cell is placed on a cell carrier to simulate the loading situation in the actual production process, allowing it to go through the same processes and operations as in actual production, such as moving on the production line and passing through various detection equipment. Record various data related to the loading of the standard cell, such as the operating status of the cell carrier and the feedback data of each sensor in contact with the standard cell, to generate a loading test result, which may include information such as the position accuracy of the standard cell on the carrier, the pressure distribution of the carrier on the standard cell, and the difference between the measurement data of each sensor and the standard value. Then, conduct an in-depth analysis of the loading test result to find out the deviation between the actual measurement data and the standard value. Based on these deviation situations, calculate the drift compensation coefficient to quantify the drift degree generated by the equipment or measurement system during the loading process. Finally, apply the calculated drift compensation coefficient to the actual measurement data. When measuring and detecting ordinary cells subsequently, correct the measurement results according to the drift compensation coefficient to make the measurement data closer to the true value. In addition, it also includes the management of the entire measurement calibration process, for example, recording information such as the time of each calibration, the specific value of the drift compensation coefficient, and the change of the measurement data before and after calibration, so as to track and evaluate the performance of the equipment and the stability of the measurement system, thereby ensuring the accuracy of cell size measurement and production quality.

[0053] Further, step S400 further includes step S430, which is to identify size anomalies based on the cell size measurement result and establish a size anomaly identification warning; step S440, which is to manage the abnormal reporting according to the size anomaly identification warning.

[0054] Preferably, according to the cell size design and quality requirements, formulate the standard range of cell size, including the allowable tolerances of key dimensions such as length, width, and height. For example, the standard length of a certain type of cell is 50mm, and the allowable tolerance range is ±0.2mm, that is, the length of qualified products should be between 49.8mm and 50.2mm. Then, conduct size anomaly identification and warning, that is, compare the collected cell size measurement results with the pre-set size standard. If the measurement result exceeds the standard range, the system determines that the cell has a size anomaly and establishes a size anomaly warning mechanism. When receiving the size anomaly identification warning, the system will record in detail the relevant information of the abnormal cell, including but not limited to the unique identifier of the cell (such as the serial number), production batch, production time, specific abnormal size value, and the deviation degree from the standard value. Finally, conduct abnormal reporting management to ensure that size anomalies are identified and processed in a timely manner.

[0055] In the above text, with reference to Figure 1A method for measuring the size of a battery cell based on machine vision according to an embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe a machine vision-based battery cell size measurement system according to an embodiment of the present invention.

[0056] The machine vision-based battery cell size measurement system according to an embodiment of the present invention is used to solve the technical problems in the prior art that it is difficult to capture the dynamic deformation characteristics of the battery cell under real charge and discharge cycle conditions, resulting in insufficient data fusion accuracy and real-time performance, and poor size measurement accuracy, and achieves the technical effect of improving the accuracy of battery cell size measurement and the safety of battery management. As Figure 2 shown, the machine vision-based battery cell size measurement system includes: an initial point cloud set construction module 10, a clustering deformation field construction module 20, a monitoring deformation data set generation module 30, and a battery cell size measurement result generation module 40.

[0057] The initial point cloud set construction module 10 is used to call a 3D vision module to capture the 3D point cloud of the battery cell and construct an initial point cloud set; the clustering deformation field construction module 20 is used to establish a typical working condition of the battery cell, obtain the battery cell deformation data under the typical working condition, perform deformation analysis on the battery cell deformation data, and construct a clustering deformation field, where the clustering deformation field includes a high strain region, a transition region, and a low strain region; the monitoring deformation data set generation module 30 is used to perform optimal placement of strain gauges in the clustering deformation field, establish an optimal placement result, configure the strain gauges with the optimal placement result, perform battery cell monitoring, and generate a monitoring deformation data set. The objective function of the optimal placement includes a monitoring accuracy function, a cost constraint function, and a safety constraint function; the battery cell size measurement result generation module 40 is used to perform registration on the initial point cloud set, perform contour recognition, establish contour position coordinates, perform coordinate compensation on the contour position coordinates based on the monitoring deformation data set, and generate a battery cell size measurement result according to the coordinate compensation result.

[0058] Next, the specific configuration of the monitoring deformation data set generation module 30 will be described in detail. The monitoring deformation data set generation module 30 further includes: establishing an objective function for optimal placement, and performing optimal placement according to the objective function. The objective function is as follows: ; wherein, represents the objective function, represents the strain gauge set, represents the strain gauge set the number of strain gauges in, represents based on the strain gauge set the full-field deformation reconstruction error under the scheme, represents the maximum allowable reconstruction error, the upper limit of the number of strain gauges, Characterize the safety constraint function 、 、 are the weight coefficients of the monitoring precision term, the cost constraint term, and the safety constraint term, respectively.

[0059] Next, the specific configuration of the monitoring deformation data set generation module 30 will be further described in detail. The monitoring deformation data set generation module 30 further includes: The monitoring precision function is as follows: ; Wherein is the total number of discrete grid points on the surface of the battery cell, represents any grid point, represents the th true deformation amount of the grid point, represents the deformation amount inverted from the strain data at the position of the strain gauge set .

[0060] Next, the specific configuration of the monitoring deformation data set generation module 30 will be further described in detail. The monitoring deformation data set generation module 30 further includes: configuring the initial distribution of the strain gauges through historical experience data to establish an initial distribution set, where each subset in the initial distribution set represents a distribution scheme of a set of strain gauges; performing fitness analysis of the strain gauges under the initial distribution set according to the objective function to generate a fitness analysis result; performing subset region fitness comparison within the initial distribution set according to the regional division of the clustering deformation field, and generating a regional optimization scheme according to the regional fitness comparison result and the fitness analysis result; performing iterative update of the initial distribution set according to the regional optimization scheme to complete the layout optimization of the strain gauges.

[0061] Next, the specific configuration of the monitoring deformation data set generation module 30 will be further described in detail. The monitoring deformation data set generation module 30 further includes: establishing a benchmark target value for optimization according to the fitness analysis result; performing trust superposition according to the regional fitness comparison result and the benchmark target value, and determining an optimization target according to the trust superposition result; performing update analysis of the initial distribution set with the optimization target to establish a regional optimization scheme.

[0062] Next, the specific configuration of the battery cell size measurement result generation module 40 will be described in detail. The battery cell size measurement result generation module 40 further includes: activating a laser sensor, performing laser data acquisition of the battery cell, and establishing an additional data set; after extracting the key point position coordinates according to the additional data set, performing data registration of the initial point cloud data based on the key point position coordinates, and performing contour recognition according to the data registration result to establish contour position coordinates.

[0063] Next, the specific configuration of the battery cell size measurement result generation module 40 will be further described in detail. The battery cell size measurement result generation module 40 further includes: the 3D vision module and the laser sensor are installed at the production line station, and the strain gauges are integrated into the battery cell carrier.

[0064] Next, the specific configuration of the battery cell size measurement result generation module 40 will be further described in detail. The battery cell size measurement result generation module 40 further includes: configuring a preset cycle range, and when the time node triggers the preset cycle range, performing cycle adaptive calibration, where the cycle adaptive calibration includes using a standard battery cell for a loading test to generate a loading test result, calculating a drift compensation coefficient according to the loading test result, and performing measurement calibration management according to the drift compensation coefficient.

[0065] Next, the specific configuration of the battery cell size measurement result generation module 40 will be further described in detail. The battery cell size measurement result generation module 40 further includes: using the battery cell size measurement result to identify size anomalies and establishing a size anomaly identification warning; performing anomaly reporting management according to the size anomaly identification warning.

[0066] The battery cell size measurement system based on machine vision provided by the embodiments of the present invention can execute the battery cell size measurement method based on machine vision provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0067] Although various references are made to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The various units and modules included are only divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0068] The above specific embodiments do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for measuring the size of an electric core based on machine vision, characterized in that The method includes: Invoking a 3D vision module to capture the 3D point cloud of the battery cell and constructing an initial point cloud set; Establishing typical working conditions of the battery cell, obtaining the deformation data of the battery cell under the typical working conditions, performing deformation analysis on the deformation data of the battery cell, and constructing a clustering deformation field, where the clustering deformation field includes a high-strain region, a transition region, and a low-strain region; Performing optimization for the layout of strain gauges in the clustering deformation field, establishing an optimization result for the layout, and configuring strain gauges according to the optimization result for the layout, performing battery cell monitoring, generating a monitored deformation data set, and the objective function for the layout optimization includes a monitoring accuracy function, a cost constraint function, and a safety constraint function; After registering the initial point cloud set, performing contour recognition, establishing contour position coordinates, performing coordinate compensation for the contour position coordinates based on the monitored deformation data set, and generating a battery cell size measurement result according to the coordinate compensation result.

2. The method for measuring the size of the battery cell based on machine vision according to claim 1, wherein, The performing optimization for the layout of strain gauges in the clustering deformation field includes: Establishing an objective function for the layout optimization, and performing layout optimization according to the objective function, and the objective function is as follows: ； Among them, represents the objective function, represents the strain gauge set, represents the strain gauge set the number of strain gauges within, represents based on the strain gauge set the full-field deformation reconstruction error under the scheme, represents the maximum allowable reconstruction error, the upper limit of the number of strain gauges, represents the safety constraint function, , , are the weight coefficients of the monitoring precision term, cost constraint term, and safety constraint term, respectively.

3. The method for measuring the size of the battery cell based on machine vision according to claim 2, wherein, The monitoring accuracy function is as follows: ； Among them, is the total number of discrete grid points on the surface of the battery cell, represents any grid point, represents the true deformation of the nth grid point, represents the deformation amount inversed from the strain data at the position of the strain gauge set.

4. The method for measuring the size of an electric core based on machine vision according to claim 2, wherein After establishing the objective function for the layout optimization, it includes: Configuring the initial distribution of strain gauges through historical experience data, establishing an initial distribution set, and each subset in the initial distribution set represents a distribution scheme of a set of strain gauges; Performing fitness analysis of strain gauges under the initial distribution set according to the objective function, and generating a fitness analysis result; Performing comparison of subset region fitness within the initial distribution set according to the region division of the clustering deformation field, and generating a region optimization scheme according to the region fitness comparison result and the fitness analysis result; Performing iterative update of the initial distribution set according to the region optimization scheme to complete the optimization for the layout of strain gauges.

5. The method for measuring the size of an electric core based on machine vision according to claim 4, wherein The generating a region optimization scheme according to the region fitness comparison result and the fitness analysis result includes: Establishing a benchmark target value for optimization according to the fitness analysis result; Performing trust superposition according to the region fitness comparison result and the benchmark target value, and determining an optimization target according to the trust superposition result; Performing update analysis of the initial distribution set with the optimization target to establish a region optimization scheme.

6. The method for measuring the size of an electric core based on machine vision according to claim 1, wherein After registering the initial point cloud set, performing contour recognition, establishing contour position coordinates, includes: Activating a laser sensor, performing laser data acquisition of the battery cell, and establishing an additional data set; After extracting the key point position coordinates according to the additional data set, performing data registration of the initial point cloud set based on the key point position coordinates, and performing contour recognition according to the data registration result to establish contour position coordinates.

7. The method for measuring the size of an electric core based on machine vision according to claim 6, wherein The 3D vision module and the laser sensor are installed at the production line station, and the strain gauges are integrated into the battery cell carrier.

8. The method for measuring the size of an electric core based on machine vision according to claim 1, wherein, After generating a battery cell size measurement result according to the coordinate compensation result, it includes: Configuring a preset cycle range, and when a time node triggers the preset cycle range, performing periodic adaptive calibration, where the periodic adaptive calibration includes performing a loading test using a standard battery cell, generating a loading test result, calculating a drift compensation coefficient according to the loading test result, and performing measurement calibration management according to the drift compensation coefficient.

9. The method for measuring the size of an electric core based on machine vision according to claim 1, wherein, Generating the cell size measurement result according to the coordinate compensation result further includes: Identifying size anomalies based on the cell size measurement result and establishing a size anomaly identification warning; Conducting anomaly reporting management according to the size anomaly identification warning.

10. The cell size measurement system based on machine vision is characterized in that, The system is used to implement the machine vision-based cell size measurement method according to any one of claims 1 to 9, and the system includes: An initial point cloud set construction module, configured to call a 3D vision module to capture the 3D point cloud of the cell and construct an initial point cloud set; A clustering deformation field construction module, configured to establish typical working conditions of the cell, obtain the cell deformation data under the typical working conditions, conduct deformation analysis on the cell deformation data, and construct a clustering deformation field, where the clustering deformation field includes a high strain region, a transition region, and a low strain region; A monitoring deformation data set generation module, configured to conduct optimal placement search of strain gauges in the clustering deformation field, establish an optimal placement search result, configure strain gauges according to the optimal placement search result, execute cell monitoring, generate a monitoring deformation data set, and the objective function of the optimal placement search includes a monitoring accuracy function, a cost constraint function, and a safety constraint function; A cell size measurement result generation module, configured to perform registration on the initial point cloud set, execute contour recognition, establish contour position coordinates, perform coordinate compensation on the contour position coordinates based on the monitoring deformation data set, and generate a cell size measurement result according to the coordinate compensation result.

Images