Ceramic diaphragm automatic positioning system based on high-precision visual feedback and closed-loop control

Abstract

The embodiment of the invention provides a ceramic diaphragm automatic positioning system based on high-precision visual feedback and closed-loop control, and the system comprises a parameter reference setting module which is used for setting a theoretical reference point and a visual detection parameter; the real-time visual detection module is used for collecting an image of the to-be-stacked film and outputting an image coordinate of the mark point; the core operation module is used for converting the image coordinates of the mark points into actual coordinates based on the visual detection parameters, reading a theoretical reference point of the current to-be-stacked diaphragm and calculating a position error vector between the actual coordinates and the theoretical reference point, and the position error vector comprises a translation compensation amount and a rotation compensation amount; and the motion control and closed loop feedback module is used for receiving the position error vector and generating a control instruction to drive the workbench to move. According to the system, through parameterized reference setting, real-time visual detection, intelligent error calculation and real-time motion compensation, the technical limitation is effectively overcome, and the lamination alignment precision and efficiency are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of ceramic diaphragm stacking and positioning, and in particular to an automatic ceramic diaphragm positioning system based on high-precision visual feedback and closed-loop control. Background Technology

[0002] In the manufacturing process of multilayer ceramic capacitors (MLCCs), the alignment of precision ceramic diaphragms is a critical step determining product performance and reliability. For a long time, positioning technologies in this field have generally suffered from insufficient accuracy, low efficiency, and poor consistency, making it difficult to meet the high-precision requirements of multilayer stacking. Due to the lack of a reliable unified benchmark method and real-time feedback mechanism, single-layer alignment errors tend to accumulate layer by layer during the stacking process, ultimately affecting the overall stacking quality. Furthermore, traditional methods heavily rely on manual experience for benchmark setting and parameter adjustment, which not only limits production efficiency but also presents challenges in flexibility and adaptability during product changes or batch transitions. Therefore, there is an urgent need for an automated positioning system that can achieve high precision, high stability, and adaptive capabilities to improve the overall process level and product yield of MLCC production. Summary of the Invention

[0003] This application provides an automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control. This system effectively overcomes the above-mentioned technical limitations and significantly improves the accuracy and efficiency of lamination alignment through parameterized benchmark setting, real-time visual detection, intelligent error calculation and real-time motion compensation.

[0004] In a first aspect, an automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control is provided, the system comprising:

[0005] The parameter reference setting module is used to set theoretical reference points and visual inspection parameters; the real-time visual inspection module is used to acquire images of the films to be stacked and output the image coordinates of the marked points.

[0006] The core computing module converts the image coordinates of the marked points into actual coordinates based on the visual detection parameters, reads the theoretical reference point of the current film to be stacked, and calculates the position error vector between the actual coordinates and the theoretical reference point. The position error vector includes translation compensation and rotation compensation.

[0007] The motion control and closed-loop feedback module is used to receive the position error vector and generate control commands to drive the table movement.

[0008] It should be understood that by setting theoretical benchmark points and visual inspection parameters through the parameter benchmark setting module, combined with the high-precision image acquisition and mark recognition of the real-time visual inspection module, and the coordinate transformation and error calculation of the core computing module, the motion control and closed-loop feedback module finally achieves precise motion compensation, which effectively solves the problems of product consistency and yield caused by low alignment accuracy and manual intervention in the ceramic diaphragm stacking process, and significantly improves positioning accuracy and production efficiency.

[0009] It should be understood that the markings on the ceramic diaphragm are precision alignment features pre-made in the blank area at the edge of the diaphragm during its manufacturing process through printing or stamping, in sync with the internal electrode pattern. They do not have electrical functions themselves, but are designed to provide a high-precision and high-reliability identification reference for the visual positioning system, so as to ensure that the electrode pattern on the diaphragm can be quickly and accurately positioned and aligned during the lamination process.

[0010] It should also be understood that the visual detection parameters include image preprocessing parameters, region of interest range, feature extraction algorithm type and its threshold setting, subpixel localization algorithm selection, and matching template parameters. These parameters work together to accurately identify and locate marker points from the acquired images, ensuring that the system can adapt to the differences in image features of different batches of films and meet the stability, repeatability, and anti-interference capabilities required for high-precision localization.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, the method for obtaining the theoretical reference point includes: automatically calculating and setting the average reference position as the coordinates of the theoretical reference point of the ceramic diaphragm in that batch by analyzing the images of multiple samples of the ceramic diaphragm in the same batch.

[0012] It should be understood that the automatic benchmark learning mode based on the analysis of multiple samples in the same batch effectively eliminates the subjective bias and operational differences when manually setting benchmarks. By calculating the average benchmark position, the objectivity and consistency of the theoretical benchmark coordinates are significantly improved, thus providing a more reliable and unified benchmark basis for subsequent visual positioning and error compensation. This fundamentally enhances the overall positioning accuracy of the system and the quality stability between different batches of products.

[0013] In conjunction with the first aspect, in some implementations of the first aspect, the real-time visual detection module uses a sub-pixel positioning algorithm to identify the center of the marker point, thereby obtaining the image coordinates. The sub-pixel positioning algorithm is either the gray-scale moment method or the interpolation method.

[0014] In conjunction with the first aspect, in some implementations of the first aspect, the method for obtaining the position error vector includes:

[0015] Obtain the actual coordinates of at least two of the marked points;

[0016] Calculate the translational deviation of each of the marked points from the theoretical reference point;

[0017] Based on the translational deviation of the at least two marked points, the translational compensation amount and the rotational compensation amount are obtained.

[0018] In conjunction with the first aspect, in some implementations of the first aspect, the method for obtaining the position error vector further includes:

[0019] The translation compensation amount is determined by the average of the translation deviations of at least two of the marked points;

[0020] The rotation compensation amount is determined by the angle difference between the actual line connecting the two marked points and the theoretical line connecting their corresponding theoretical reference points.

[0021] It should be understood that by obtaining the actual coordinates of at least two marker points and calculating their translational deviations from the theoretical reference point, the translational and rotational compensation amounts are comprehensively calculated based on the multi-point deviation information. The translational compensation amount is determined by the average value of the multi-point deviations to suppress random errors, while the rotational compensation amount is accurately calculated by the angle difference between the actual and theoretical lines. This significantly improves the calculation accuracy and anti-interference capability of the position error vector under complex working conditions, effectively overcomes the limitations of single-point positioning, and ensures the accuracy of overall alignment during the stacking process.

[0022] In conjunction with the first aspect, in some implementations of the first aspect, after the motion control and closed-loop feedback module drives the worktable to move, it triggers the real-time vision detection module to re-detect, and the core computing module calculates the remaining error based on the re-detection result and makes adjustments until the remaining error is lower than the preset tolerance. It should be understood that the remaining error refers to the slight deviation between the actual coordinates of the marker point and the theoretical reference coordinates obtained by the real-time vision detection module after the motion control and closed-loop feedback module drives the worktable to perform the initial position error compensation; while the preset tolerance is the maximum permissible error threshold preset according to the precision requirements of the multilayer ceramic capacitor stacking process. The system provided in this application continuously reduces the remaining error until it is lower than the upper limit of this tolerance through iterative detection and compensation control, thereby ensuring that the final positioning accuracy meets production requirements.

[0023] In conjunction with the first aspect, in some implementations of the first aspect, the real-time visual inspection module includes at least two industrial cameras, which are aimed at the worktable at different angles to synchronously acquire images of the films to be stacked.

[0024] In conjunction with the first aspect, in some implementations of the first aspect, the motion control and closed-loop feedback module includes a high-precision force sensor, which is integrated on the actuator of the worktable, and is used to monitor microscopic mechanical signals during the stacking process in real time.

[0025] It should be understood that by introducing a multi-angle industrial camera synchronous acquisition system, multi-view image information of the diaphragm can be effectively acquired, significantly improving the ability to identify marker points and adaptability to complex working conditions. At the same time, the integration of high-precision force sensors to monitor the stacking mechanical signals in real time enables the system to sense and respond to the interlayer contact state and minute deformations, thereby further enhancing the physical coupling quality control of the stacking process on the basis of visual positioning, and comprehensively improving positioning accuracy, product consistency and yield. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of an automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control, provided as an embodiment of this application. Detailed Implementation

[0027] The terminology used in the following embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of this application, “at least one” and “one or more” refer to one, two, or more than two. The term “and / or” is used to describe the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can indicate: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0028] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0029] In the manufacturing of multilayer ceramic capacitors (MLCCs), the alignment of the ceramic diaphragms directly determines the product's performance and reliability. Existing positioning technologies suffer from insufficient accuracy, low efficiency, and poor consistency, making it difficult to meet micron-level or even sub-meter-level stacking accuracy requirements. Due to the lack of a unified benchmark and real-time feedback mechanism, single-layer alignment errors accumulate layer by layer, severely impacting overall quality. Furthermore, traditional methods rely on manual experience for benchmark setting and parameter adjustment, resulting in low production efficiency and insufficient flexibility during model or batch changes. Therefore, a high-precision, highly stable, and adaptive automatic positioning system is urgently needed to improve MLCC production yield and process efficiency.

[0030] This application provides an automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control, which can effectively overcome the above-mentioned problems.

[0031] The technical solutions provided in the embodiments of this application will be described below with reference to the accompanying drawings.

[0032] Figure 1 This is a schematic diagram of an automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control, provided as an embodiment of this application.

[0033] refer to Figure 1 In some examples, the system includes:

[0034] The parameter reference setting module is used to set theoretical reference points and visual inspection parameters; the real-time visual inspection module is used to acquire images of the films to be stacked and output the image coordinates of the marked points.

[0035] The core computing module converts the image coordinates of the marked points into actual coordinates based on the visual detection parameters, reads the theoretical reference point of the current film to be stacked, and calculates the position error vector between the actual coordinates and the theoretical reference point. The position error vector includes translation compensation and rotation compensation.

[0036] The motion control and closed-loop feedback module is used to receive the position error vector and generate control commands to drive the table movement.

[0037] In some examples, the theoretical benchmark points are obtained in the following ways:

[0038] By analyzing the images of ceramic membranes from multiple samples in the same batch, the average reference position is automatically calculated and set as the theoretical reference point coordinates for that batch of ceramic membranes.

[0039] In some examples, the real-time visual detection module uses a subpixel localization algorithm to identify the center of the marker point, thereby obtaining the image coordinates. The subpixel localization algorithm is either the gray-scale moment method or the interpolation method.

[0040] In one possible implementation, the real-time visual detection module first enhances the contrast between the marker point and the background through image preprocessing. Then, it uses a gray-scale moment subpixel localization algorithm within a preset region of interest to accurately determine the center position of the marker point by calculating the centroid coordinates of the gray-scale distribution in the edge region of the marker point. Finally, it outputs image coordinates with subpixel accuracy. This method effectively overcomes the accuracy limitations of traditional pixel-level localization and significantly improves the localization accuracy and stability of the system under complex imaging conditions.

[0041] In some examples, the method for obtaining the position error vector includes:

[0042] Obtain the actual coordinates of at least two of the marked points;

[0043] Calculate the translational deviation of each of the marked points from the theoretical reference point;

[0044] Based on the translational deviation of the at least two marker points, the translational compensation amount and the rotational compensation amount are obtained. In some examples, the method for obtaining the position error vector further includes:

[0045] The translation compensation amount is determined by the average of the translation deviations of the at least two of the marked points; the rotation compensation amount is determined by the angle difference between the actual line connecting the two marked points and the theoretical line connecting their corresponding theoretical reference points.

[0046] In one possible implementation, the system first obtains the coordinates of two marker points on the current diaphragm in the real-world coordinate system through a calibrated vision system, and simultaneously reads the coordinates of the corresponding theoretical reference points from the storage unit. Then, the core computing module calculates the translational deviations of the two marker points in the X and Y axes, respectively. Next, the arithmetic mean of the translational deviations of the two marker points is taken as the final translational compensation. At the same time, the rotational compensation is obtained by calculating the angle difference between the line connecting the two actual marker points and the line connecting the corresponding theoretical reference points. Finally, a position error vector containing the translational and rotational compensation is generated, providing precise compensation commands for the motion control system.

[0047] In some examples, after the motion control and closed-loop feedback module drives the worktable to move, it triggers the real-time vision detection module to re-detect, and the core computing module calculates the remaining error based on the re-detection result and makes adjustments until the remaining error is lower than the preset tolerance.

[0048] In some examples, the real-time visual inspection module includes at least two industrial cameras that are aimed at the worktable at different angles to synchronously acquire images of the films to be stacked.

[0049] In some examples, the motion control and closed-loop feedback module includes a high-precision force sensor integrated on the actuator of the worktable, which is used to monitor micromechanical signals during the stacking process in real time.

[0050] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or variations made by those skilled in the art based on the content disclosed in the present invention should be included within the scope of protection set forth in the claims.

Claims

1. An automatic positioning system for ceramic diaphragms based on high-precision visual feedback and closed-loop control, characterized in that, The system includes: The parameter reference setting module is used to set theoretical reference points and visual inspection parameters; the real-time visual inspection module is used to acquire images of the films to be stacked and output the image coordinates of the marked points. The core computing module converts the image coordinates of the marked points into actual coordinates based on the visual detection parameters, reads the theoretical reference point of the current film to be stacked, and calculates the position error vector between the actual coordinates and the theoretical reference point. The position error vector includes translation compensation and rotation compensation. The motion control and closed-loop feedback module is used to receive the position error vector and generate control commands to drive the table movement.

2. The system according to claim 1, characterized in that, The method for obtaining the theoretical reference point includes: by analyzing the images of multiple ceramic membranes from the same batch, automatically calculating and setting the average reference position as the coordinates of the theoretical reference point of the ceramic membranes in that batch.

3. The system according to claim 2, characterized in that, The real-time visual detection module uses a subpixel localization algorithm to identify the center of the marker point, thereby obtaining the image coordinates. The subpixel localization algorithm is either the gray-scale moment method or the interpolation method.

4. The system according to claim 3, characterized in that, The method for obtaining the position error vector includes: Obtain the actual coordinates of at least two of the marked points; Calculate the translational deviation of each of the marked points from the theoretical reference point; Based on the translational deviation of the at least two marked points, the translational compensation amount and the rotational compensation amount are obtained.

5. The system according to claim 4, characterized in that, The method for obtaining the position error vector further includes: The translation compensation amount is determined by the average of the translation deviations of the at least two of the marked points; The rotation compensation amount is determined by the angle difference between the actual line connecting the two marked points and the theoretical line connecting their corresponding theoretical reference points.

6. The system according to claim 5, characterized in that, After the motion control and closed-loop feedback module drives the worktable to move, it triggers the real-time vision detection module to re-detect, and the core computing module calculates the remaining error based on the re-detection result and makes adjustments until the remaining error is lower than the preset tolerance.

7. The system according to claim 6, characterized in that, The real-time visual inspection module includes at least two industrial cameras, which are aimed at the worktable at different angles to synchronously acquire images of the films to be stacked.

8. The system according to claim 7, characterized in that, The motion control and closed-loop feedback module includes a high-precision force sensor, which is integrated on the actuator of the worktable. The high-precision force sensor is used to monitor the micro-mechanical signals during the stacking process in real time.

Images