A ceramic substrate warpage detection method and system based on laser three-dimensional point cloud

Abstract

This application relates to a method and system for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds. It pertains to the technical field of ceramic substrate warpage detection, including acquiring ceramic substrate contact height and point cloud detection data; analyzing the ceramic substrate contact height and point cloud detection data to determine abnormal lifting block results; determining whether the abnormal lifting block results meet preset abnormal lifting result requirements; if so, acquiring the warpage compensation height; analyzing the warpage compensation height to calibrate a preset warpage detection device and determine the ceramic substrate warpage; if not, performing data analysis on the point cloud detection data to determine the ceramic substrate warpage. This application improves the detection efficiency and accuracy of ceramic substrates.

Description

Technical Field

[0001] This application relates to the technical field of ceramic substrate warpage detection, and in particular to a method and system for ceramic substrate warpage detection based on laser three-dimensional point cloud. Background Technology

[0002] Ceramic substrate warpage detection refers to the process of scanning the laser three-dimensional point cloud data of a ceramic substrate using a detection device, and then analyzing the point cloud data to determine the warpage of the ceramic substrate.

[0003] In related technologies, when detecting the warpage of ceramic substrates, a laser point cloud detection device is usually used for direct detection. During the detection process, the feeding mechanism fixes the ceramic substrates one by one on the reference lifting support platform. After the fixing is completed, the three-dimensional optical scanning unit acquires the point cloud data of the ceramic substrates, and then the data processing module analyzes the point cloud data to determine the warpage of the ceramic substrates.

[0004] Regarding the aforementioned technologies, when using a detection device to directly detect the warpage of ceramic substrates, the detection efficiency is low because each ceramic substrate is inspected individually. Furthermore, direct detection can lead to discrepancies between the reference lifting platform and the reference surface, or interference from local debris, causing deviations in the reference plane for point cloud detection of the ceramic substrate. This results in low accuracy in detecting the warpage of the ceramic substrate, leaving room for improvement. Summary of the Invention

[0005] To improve the detection efficiency and accuracy of ceramic substrates, this application provides a method and system for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds.

[0006] Firstly, this application provides a method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds, employing the following technical solution: A method for detecting the warpage of ceramic substrates based on laser 3D point clouds includes: Acquire ceramic sheet contact height and point cloud detection data; Analyze the contact height of the ceramic sheet and the point cloud detection data to determine the abnormal results of the lifting block; Determine whether the abnormal result of the lifting block meets the preset requirements for abnormal lifting results; If so, obtain the warp compensation height; The warpage compensation height is analyzed to calibrate the preset warpage detection device and determine the warpage of the ceramic sheet; If not, perform data analysis on the point cloud detection data to determine the warpage of the ceramic sheet.

[0007] Optionally, the steps of analyzing the ceramic sheet contact height and point cloud detection data to determine the abnormal results of the lifting block include: The contact height of the ceramic sheet and the point cloud detection data were analyzed to determine the compensation contact height; Calculate the absolute deviation between the compensated contact height and the preset unbiased contact height to determine the compensation height difference; Determine whether the compensation height difference is greater than the preset upper limit threshold for deviation; If it is greater than the preset abnormal result, then the abnormal result will be determined as the abnormal result of the lifting block; If the result is not greater than the preset no-abnormal result, then the lifting block abnormal result will be determined.

[0008] Optionally, the steps of analyzing the ceramic sheet contact height and point cloud detection data to determine the compensating contact height include: Plane fitting is performed on the point cloud detection data to determine the plane fitting equation; Obtain the coordinate range of the lifting block; Data extraction is performed on the point cloud detection data based on the coordinate range of the lifting block to determine the contact point cloud of the lifting block; Substitute the contact point cloud of the lifting block into the plane fitting equation to determine the reference fitting height; Calculate the height difference between the contact point cloud of the lifting block and the reference fitted height to determine the directional distance of the point cloud; Numerical analysis of the directed distance in the point cloud is performed to determine the directed distance quantiles; Calculate the difference between the ceramic tile contact height and the directional distance quantile to determine the compensation contact height.

[0009] Optionally, the steps of analyzing the warpage compensation height to calibrate the preset warpage detection device and determine the warpage of the ceramic sheet include: Linear fitting is performed on the warpage compensation height to determine the goodness of linear fit and the stroke-fitted straight line; Data is extracted from the stroke-fitted straight line to determine the slope of the linear fit; Determine whether the goodness of linear fit and the slope of linear fit meet the preset requirements for cylinder asynchrony results; If so, the warpage detection device is calibrated based on the warpage compensation height and the stroke fitting straight line. If not, obtain the no-load test result; The warpage detection device is calibrated based on the no-load test results; Obtain the deviation lifting block number; The warpage of the ceramic sheet is determined based on the deviation lifting block number.

[0010] Optionally, the step of calibrating the warpage detection device based on the warpage compensation height and stroke fitting straight line includes: Obtain the coordinates of the left and right cylinders; Input the coordinates of the left and right cylinders into the stroke fitting line to determine the cylinder stroke deviation; Calculate the average warpage compensation height to determine the average compensation contact value; The cylinder stroke deviation, the average compensation contact value, and the preset reference stroke speed are input into the preset speed compensation model to determine the stroke compensation speed. The warpage detection device is calibrated based on the travel compensation speed.

[0011] Optionally, the step of calibrating the warpage detection device based on the no-load test results includes: Linear fitting is performed on the unloaded test results to determine the top surface reference equation; Based on the no-load test results, determine the maximum and minimum fitted height differences in the top surface reference equation; Calculate the difference between the maximum and minimum fitted height differences to determine the flatness of the lifting block; The warpage detection device is calibrated based on the flatness of the lifting block and the difference between the maximum fitted height.

[0012] Optionally, the step of calibrating the warpage detection device based on the flatness of the lifting block and the maximum fitted height difference includes: Calculate the quotient of the maximum fitted height difference and the flatness of the lifting block to determine the lifting block protrusion coefficient; Determine whether the jacking block protrusion coefficient, maximum fitting height difference, and jacking block flatness meet the preset requirements for local debris results; If so, the preset cleaning device will be activated to clean and calibrate the warpage detection device; If not, the warpage detection device should be replaced and calibrated.

[0013] Optionally, the step of determining the warpage of the ceramic sheet based on the deviation lifting block number includes: Obtain historical warp measurements; Numerical analysis was performed on historical warpage to determine the historical warpage mean and historical warpage variance; The warpage of the deviation station is obtained based on the deviation lifting block number and the preset flow waiting time; Numerical analysis was performed on the warpage of the deviation workstation to determine the mean and variance of the deviation warpage. Calculate the absolute deviation of the historical warp mean and the deviation warp mean to determine the warp mean deviation; Calculate the absolute deviation of the historical warp variance and the bias warp variance to determine the warp variance bias; Determine whether the mean warpage deviation and the variance warpage deviation meet the preset requirements for normal flow results; If so, the preset ceramic sheet will be re-inspected to determine the degree of warpage of the ceramic sheet; If not, the ceramic sheet will be retrospectively inspected to determine its warpage.

[0014] Secondly, this application provides a ceramic substrate warpage detection system based on laser three-dimensional point clouds, employing the following technical solution: A ceramic substrate warpage detection system based on laser three-dimensional point cloud includes: The acquisition module is used to acquire the contact height of the ceramic sheet, point cloud detection data, and warpage compensation height. A memory for storing a program for a method of detecting the warpage of a ceramic substrate based on laser three-dimensional point cloud as described in any of the above claims; The processor and the program in the memory can be loaded and executed by the processor to implement a method for detecting the warpage of a ceramic substrate based on laser three-dimensional point cloud as described in any of the above.

[0015] In summary, this application includes at least one of the following beneficial technical effects: 1. By analyzing the contact height of the ceramic sheet and the point cloud detection data, the abnormal lifting situation of the lifting block is determined. When the lifting block exhibits abnormal lifting, the warpage detection device is calibrated, and the warpage of the ceramic sheet is determined again. When the lifting block does not exhibit abnormal lifting, the warpage of the ceramic sheet is determined directly by analyzing the point cloud detection data, thereby improving the detection efficiency of the ceramic substrate. 2. By analyzing the point cloud monitoring data of the ceramic substrate, the reference fitting height is determined. Then, the contact height of the ceramic sheet is compensated according to the reference fitting height to determine the compensated contact height when the ceramic sheet contacts the lifting block. This helps to determine whether the lifting block has abnormal lifting conditions, thereby improving the accuracy of judging abnormal lifting conditions of the lifting block. 3. By analyzing historical warp values, the mean and variance of historical warp values ​​are determined, and the warp value of the deviation station in the buffer area is obtained. Based on the mean deviation and variance deviation between historical warp values ​​and deviation station warp values, it is determined whether the ceramic sheet needs to be re-inspected, thereby improving the detection accuracy of ceramic substrates. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for detecting the warpage of a ceramic substrate based on laser three-dimensional point cloud in an embodiment of this application.

[0017] Figure 2 This is a flowchart illustrating the analysis of ceramic sheet contact height and point cloud detection data in this application embodiment to determine abnormal results of the lifting block.

[0018] Figure 3This is a flowchart illustrating the analysis of ceramic sheet contact height and point cloud detection data in this embodiment of the application to determine the compensation contact height.

[0019] Figure 4 This is a flowchart illustrating the process of analyzing the warpage compensation height in this embodiment of the application to calibrate a preset warpage detection device and determine the warpage of the ceramic sheet.

[0020] Figure 5 This is a flowchart of the warp detection device that calibrates the warp compensation height and stroke fitting straight line according to the embodiments of this application.

[0021] Figure 6 This is a flowchart of the process for calibrating the warpage detection device based on the no-load detection results in the embodiments of this application.

[0022] Figure 7 This is a flowchart of the warpage detection device calibrated based on the difference between the flatness of the lifting block and the maximum fitted height in the embodiments of this application.

[0023] Figure 8 This is a flowchart illustrating the determination of ceramic sheet warpage based on the deviation lifting block number in an embodiment of this application. Detailed Implementation

[0024] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 8 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0025] This application discloses a method and system for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds. Specifically, it discloses a processing terminal and a warpage detection device. The processing terminal acquires the contact height of the ceramic substrate and point cloud detection data. The contact height and point cloud detection data are analyzed to determine abnormal lifting conditions of the lifting block. When abnormal lifting conditions occur, the warpage detection device is calibrated, and the warpage of the ceramic substrate is determined again. When no abnormal lifting conditions occur, the point cloud detection data is directly analyzed to determine the warpage of the ceramic substrate, thereby improving the detection efficiency of ceramic substrates.

[0026] Reference Figure 1 This application discloses a method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds, comprising the following steps: Step S100: Obtain ceramic sheet contact height and point cloud detection data.

[0027] Among them, the ceramic sheet contact height refers to the actual lifting height when the lifting block contacts the ceramic sheet. The processing terminal retrieves the measurement values ​​of the pressure sensors installed on the top of each lifting block in real time. When the measurement data of the pressure sensor changes abruptly, that is, when the lifting block contacts the ceramic sheet, the height displacement reading of the servo electric cylinder magnetic scale is latched until the height displacement reading of all lifting blocks is recorded and determined. Then, the independent magnetic scale displacement of the four lifting blocks in a row is converted into the detection coordinate system of the warp detection device for determination.

[0028] Point cloud detection data refers to the three-dimensional laser point cloud detection data of the surface of the ceramic sheet being inspected. The processing terminal retrieves the three-dimensional laser point cloud detection data at the current moment, and then divides and extracts the three-dimensional laser point cloud detection data into blocks according to the coordinate range of the ceramic sheet calibrated in the system.

[0029] Step S101: Analyze the contact height of the ceramic sheet and the point cloud detection data to determine the abnormal results of the lifting block.

[0030] Among them, the abnormal result of the lifting block refers to the control indicator in the ceramic tile warpage detection process, including abnormal results and no abnormal results. The abnormal result indicator processing terminal needs to calibrate the warpage detection device and then re-inspect the warpage of the ceramic tile. The no abnormal result indicator processing terminal does not need to calibrate the warpage detection device. The ceramic tile warpage is determined by numerical analysis of the point cloud detection data. The processing terminal determines this by analyzing the contact height of the ceramic tile and the point cloud detection data. The specific analysis steps are as follows: Figure 2 The steps in the process.

[0031] Step S102: Determine whether the abnormal result of the lifting block meets the preset requirements for abnormal lifting results.

[0032] Among them, abnormal lifting result refers to an abnormal result of the lifting block, which is calibrated by the operator in the system and used to determine the control flag of the current system. The requirement for abnormal lifting result is that it must be consistent with the abnormal lifting result.

[0033] By processing the terminal to determine whether the abnormal result of the lifting block meets the requirements of the abnormal lifting result, it can be determined whether the warpage detection device needs to be calibrated, and then the warpage of the ceramic sheet can be re-inspected, thereby improving the detection accuracy of the ceramic sheet warpage.

[0034] Step S1021: If yes, then obtain the warp compensation height.

[0035] If the processing terminal determines that the abnormal result of the lifting block is abnormal, it indicates that the warpage detection device needs to be calibrated and the warpage of the ceramic sheet needs to be re-inspected. Therefore, the warpage compensation height is obtained to provide data support for the subsequent determination of the warpage of the ceramic sheet.

[0036] The warpage compensation height refers to the contact height of the ceramic sheet after compensating for the warpage of the ceramic sheet itself. It is determined by the processing terminal when the abnormal result of the lifting block is determined, by retrieving the compensation contact height calculated during the process of determining the abnormal result of the lifting block.

[0037] Step S1022: Analyze the warpage compensation height to calibrate the preset warpage detection device and determine the warpage of the ceramic sheet.

[0038] Among them, ceramic tile warpage refers to the warpage of the currently inspected ceramic tile, or the warpage of the currently inspected ceramic tile and the re-inspection results of the warpage of all inspected ceramic tiles during the circulation waiting period. When the lifting block abnormality result is abnormal, but the warpage of the inspected ceramic tiles during the circulation waiting period does not show a mean or variance deviation relative to the historical warpage, the ceramic tile warpage is the warpage of the currently inspected ceramic tile. Conversely, the ceramic tile warpage is the warpage of the currently inspected ceramic tile and the re-inspection results of the warpage of all inspected ceramic tiles during the circulation waiting period. The warpage detection results of the corresponding ceramic tile are updated based on the re-inspection results, thereby improving the detection accuracy of ceramic tile warpage. The processing terminal analyzes the warpage compensation height and re-inspects the warpage detection device to determine the accuracy. The specific analysis steps refer to [reference needed]. Figure 4 The steps in the process.

[0039] The warpage detection device integrates a placement robotic arm, ceramic tile carriers, a carrier conveyor line, a cylinder lifting assembly, a 3D laser point cloud scanning device, and a pickup robotic arm. At the front end of the device, the placement robotic arm places the ceramic tiles onto the ceramic tile carriers. Each carrier has a row of four ceramic tile placement grooves, with a perforated opening beneath each groove for the cylinder lifting assembly to extend. As the carrier moves along the carrier conveyor line to the position of the 3D laser point cloud scanning device, it lifts the row of four ceramic tiles, placing them on the same reference detection plane. The carrier continues moving along the conveyor line. After testing, the ceramic tiles are picked up by the pickup robotic arm and placed in a buffer area to await the next testing step. When the ceramic tile carrier moves to the position of the cylinder lifting assembly and the 3D laser point cloud scanning device, the cylinder lifting assembly located directly below the ceramic tile to be inspected starts lifting. The two symmetrical cylinders on the left and right simultaneously lift the flat base plate. Four lifting blocks are set on the flat base plate to lift the ceramic tile through the hollow openings on the carrier. After the lifting is completed and the ceramic tile is stabilized, the 3D laser point cloud scanning device performs point cloud detection on the upper surface of the ceramic tile. After the detection is completed, the cylinder lifting assembly slowly falls, causing the ceramic tile to return to the groove on the ceramic tile carrier. The carrier conveying device starts moving again, conveying the inspected ceramic tile carrier backward and conveying the uninspected ceramic tile to the inspection position. The base plate is raised again, causing the lifting blocks to lift the new ceramic tile to be inspected.

[0040] Step S1023: If not, perform data analysis on the point cloud detection data to determine the warpage of the ceramic sheet.

[0041] If the processing terminal determines that the abnormal result of the lifting block is no abnormal result, it means that the warpage detection device does not need to be calibrated at this time. Therefore, the point cloud detection data is analyzed to determine the warpage of the ceramic sheet.

[0042] Ceramic tile warpage refers to the warpage of a row of ceramic tiles at the current detection position. It is determined by the processing terminal based on the point cloud detection data, when the lifting block abnormality result is determined to be no abnormality result. The specific calculation formula is as follows: .

[0043] In the formula, For ceramic tile warpage, This represents the maximum detection height in the point cloud detection data. This represents the minimum detection height in the point cloud detection data.

[0044] Therefore, a unified absolute reference plane is determined based on the lifting surface of the equipment. The reference plane is set as the zero point of the z-axis height in the point cloud detection coordinate system. The warpage of the ceramic sheet is determined based on the height difference between the highest and lowest points in the laser point cloud.

[0045] Reference Figure 2 The steps for analyzing the contact height of the ceramic sheet and point cloud detection data to determine abnormal results of the lifting block include: Step S200: Analyze the ceramic sheet contact height and point cloud detection data to determine the compensation contact height.

[0046] The compensated contact height refers to the contact height of the ceramic sheet after compensating for the warping effect of the ceramic sheet itself. That is, if the ceramic sheet has unevenness, the contact height between the ceramic sheet and the lifting block is compensated to a planar contact state to obtain the true lifting situation of the lifting block, thus improving the recognition accuracy of abnormal lifting situations. This is determined by the processing terminal through analysis of the ceramic sheet contact height and point cloud detection data. Specific analysis steps are described in [reference needed]. Figure 3 The steps in the process.

[0047] Step S201: Calculate the absolute deviation between the compensated contact height and the preset unbiased contact height to determine the compensation height difference.

[0048] Among them, the unbiased contact height refers to the reference contact height when the lifting block contacts the unwarped ceramic sheet under the condition of no abnormal lifting. The operator first performs manual laser point cloud detection on each ceramic sheet to screen and determine the unwarped ceramic sheets. Then, the unwarped ceramic sheets are placed on the warping detection device under the condition of no bias and no abnormal lifting, which is confirmed by manual inspection, for offline lifting pre-experiment. The average value of the contact height is determined by repeated experiments, which is the unbiased contact height.

[0049] The compensation height difference refers to the absolute deviation between the compensation contact height and the unbiased contact height, that is, the deviation between the contact height between the lifting block and the reference ceramic plate during the lifting process. It is determined by the processing terminal by calculating the absolute deviation between the compensation contact height and the preset unbiased contact height.

[0050] Step S202: Determine whether the compensation height difference is greater than the preset upper limit threshold of the deviation.

[0051] The upper limit of deviation refers to the upper limit of the compensation height difference when the lifting block does not experience any lifting abnormalities, resulting in asynchronous lifting processes and uneven lifting reference surfaces. It is calibrated by the operator through offline pre-experimentation. The specific experimental steps are as follows: under the lifting abnormality-free working conditions determined by the operator, a lifting pre-experiment is conducted on a ceramic sheet without abnormal warping. After determining the contact height data, the mean and standard deviation of the contact height difference data are calculated. Then, the sum of the mean and three times the standard deviation is calculated, which is the upper limit of deviation.

[0052] By processing the terminal to determine whether there is a compensation height difference greater than the upper limit threshold of the deviation among the compensation height differences corresponding to the four lifting blocks, it can be determined whether there is an abnormal lifting situation of the lifting blocks, which leads to asynchronous lifting process and uneven lifting reference surface, thereby improving the detection accuracy of ceramic sheet warpage.

[0053] Step S2021: If it is greater than, then the preset abnormal result is determined as the abnormal result of the lifting block.

[0054] If the processing terminal determines that there is a compensation height difference greater than the upper limit threshold of the deviation among the compensation height differences corresponding to the four lifting blocks, it indicates that the lifting blocks are experiencing lifting abnormalities, resulting in asynchronous lifting processes and uneven lifting reference surfaces. In this case, the warpage detection device needs to be calibrated. Therefore, the abnormal result is recorded as an abnormal result of the lifting blocks, providing a control indicator for subsequent calibration of the warpage detection device.

[0055] Abnormal results refer to situations where the lifting blocks exhibit abnormal lifting conditions, such as asynchronous lifting or uneven reference surfaces. When it is necessary to calibrate the warpage detection device, the corresponding control indicators are calibrated by the operator in the system.

[0056] Step S2022: If it is not greater than, then determine the preset no-abnormality result as the abnormality result of the lifting block.

[0057] Among them, if it is determined by the processing terminal that the compensated height difference is not greater than the deviation upper limit threshold, it indicates that the lifting block does not have abnormal lifting at this time and the lifting process is synchronized. Therefore, at this time, the no-abnormality result is determined as the abnormality result of the lifting block, providing a control identifier for directly determining the warpage of the ceramic chip according to the point cloud detection data in the subsequent process.

[0058] The no-abnormality result refers to the control identifier corresponding to the situation where the lifting block does not have abnormal lifting, the lifting process is synchronized, and there is no need to calibrate the warpage detection data, and the warpage of the ceramic chip can be directly calculated. It is calibrated by the operator in the system.

[0059] Refer to Figure 3 , the steps for analyzing the contact height of the ceramic chip and the point cloud detection data to determine the compensated contact height include: Step S300: Perform plane fitting on the point cloud detection data to determine the plane fitting equation.

[0060] Among them, the plane fitting equation refers to the reference fitting plane of each ceramic chip on the currently inspected column, which is determined by the processing terminal through least squares fitting of the point cloud detection data.

[0061] Step S301: Obtain the coordinate range of the lifting block.

[0062] Among them, the coordinate range of the lifting block refers to the plane coordinate range of the top plane of the lifting block in the point cloud detection coordinate system of the warpage detection device, such as {(x, y)|x < a, y < a}, which is used to extract data from the plane fitting equation, extract the plane fitting equation corresponding to the contact part between the lifting block and the ceramic chip, and provide data support for determining the compensated contact height in the subsequent process. The processing terminal first retrieves the specific layout parameters of the lifting block, determines the relative position of the lifting block relative to the placement groove of the ceramic chip, and then inversely deduces the plane coordinate range of the lifting block based on the coordinates of the ceramic chip in the laser coordinate system and the relative coordinates.

[0063] Step S302: Extract data from the point cloud detection data according to the coordinate range of the lifting block to determine the contact point cloud of the lifting block.

[0064] Among them, the contact point cloud of the lifting block refers to the point cloud detection data of the ceramic chip within the contact range between the lifting block and the ceramic chip, which is determined by the processing terminal through extracting data from the point cloud detection data within the coordinate range of the lifting block according to the coordinate range of the lifting block.

[0065] Step S303: Substitute the contact point cloud of the lifting block into the plane fitting equation to determine the reference fitting height.

[0066] The reference fitting height refers to the fitting height of each plane coordinate point in the xy plane coordinates corresponding to the contact point cloud of the lifting block in the plane fitting equation. It is calculated and determined by the processing terminal by substituting the xy plane coordinates corresponding to the contact point cloud of the lifting block into the plane fitting equation.

[0067] Step S304: Calculate the height difference between the contact point cloud of the lifting block and the reference fitted height to determine the directional distance of the point cloud.

[0068] Among them, the directional distance between the point cloud and the contact point cloud of the lifting block and the reference fitting plane is a directional height distance. A positive distance indicates that the point cloud is convex, and a negative distance indicates that the point cloud is concave. It is determined by the processing terminal by calculating the height difference between the corresponding height of the contact point cloud of the lifting block and the corresponding reference fitting height.

[0069] Step S305: Perform numerical analysis on the directed distance of the point cloud to determine the directed distance quantiles.

[0070] The directed distance quantile refers to the quantile value of the directed distance in the point cloud. The processing terminal first calculates the arithmetic mean of the directed distance in the point cloud, and then analyzes the arithmetic mean. When the arithmetic mean is positive, that is, when the ceramic piece is convex, the directed distance in the point cloud corresponding to the 90th quantile position is taken as the directed distance quantile. When the arithmetic mean is negative, that is, when the ceramic piece is concave, the directed distance in the point cloud corresponding to the 10th quantile position is taken as the directed distance quantile. This eliminates the influence of extreme concavity and convexity and improves the compensation accuracy of ceramic piece warping.

[0071] Step S306: Calculate the difference between the ceramic sheet contact height and the directional distance quantile to determine the compensation contact height.

[0072] The compensation contact height is consistent with the compensation contact height in step S200, and is determined by the processing terminal by calculating the difference between the ceramic sheet contact height and the directed distance quantile after determining the directed distance quantile.

[0073] Reference Figure 4 The steps for analyzing the warpage compensation height, calibrating the preset warpage detection device, and determining the warpage of the ceramic sheet include: Step S400: Perform linear fitting on the warp compensation height to determine the goodness of fit and the stroke fitting line.

[0074] Among them, the linear fit goodness refers to the linear fit goodness of the stroke fitting line, with a range of 0-1. The closer the linear fit goodness is to 1, the stronger the linear relationship of the fitting line. That is, the warpage compensation height is caused by the tilting of the lifting base plate due to the asynchronous lifting of the cylinders, which leads to the linear contact height relationship. It is used to judge the asynchronous situation of the cylinders. It is determined by the processing terminal after determining the stroke fitting line, based on the sum of squares of the deviations between the measured height value and the fitted height value.

[0075] The stroke fitting line refers to the linear function obtained by solving the warpage compensation height and the fixed lateral coordinates of the center point of each lifting block after the four lifting blocks are uniformly converted to the absolute reference height zero point of the warpage detection device, using the least squares algorithm. The processing terminal then performs least squares linear fitting on the warpage compensation height and the coordinates of the center point of each lifting block in the detection coordinate system.

[0076] Step S401: Extract data from the stroke-fitted straight line to determine the slope of the linear fit.

[0077] Among them, the linear fitting slope refers to the slope of the stroke-fitted straight line, which is determined by the processing terminal through data extraction of the slope of the stroke-fitted straight line.

[0078] Step S402: Determine whether the goodness of linear fit and the slope of linear fit meet the preset requirements for cylinder asynchrony results.

[0079] The cylinder asynchrony result refers to a linear fit goodness of fit less than a preset asynchrony goodness threshold and an absolute value of the linear fit slope greater than a preset asynchrony slope threshold. The requirement for the cylinder asynchrony result is that it is consistent with the cylinder asynchrony result.

[0080] The asynchrony goodness threshold refers to the upper limit of the linear fit goodness corresponding to cylinder asynchrony. When the linear fit goodness is less than this threshold, there is a risk of cylinder asynchrony. It is determined by the operator through offline pre-experimentation. The specific experimental steps are as follows: The operator conducts multiple sets of pre-lifting experiments on the non-warped ceramic plate under normal lifting conditions to determine the linear fit goodness corresponding to the normal conditions. Then, the first fault-free goodness is determined based on the sum of the mean and three standard deviations of the linear fit goodness. Then, based on the cylinder asynchrony situations that have occurred in historical working conditions, the cylinder asynchrony working conditions are set based on the smallest asynchrony situation among the cylinder asynchrony situations. Multiple experiments are conducted, and the second fault goodness is determined by the sum of the mean and three standard deviations of the linear fit goodness. Finally, the mean of the first fault-free goodness and the second fault-free goodness is calculated as the asynchrony goodness threshold.

[0081] The asynchronous slope threshold refers to the lower limit of the linear fitting slope corresponding to cylinder asynchronous conditions. When the linear fitting slope is greater than this threshold, there is a risk of cylinder asynchronousness. It is determined by the operator through offline pre-experimentation. The specific experimental steps are as follows: The operator conducts multiple sets of pre-lifting experiments on the non-warped ceramic plate under normal lifting conditions to determine the corresponding linear fitting slope under normal conditions. Then, based on the sum of the mean and three standard deviations of the linear fitting slope, the first fault-free slope is determined. Then, based on the cylinder asynchronous conditions that have occurred in historical working conditions, the cylinder asynchronous working condition is set based on the smallest asynchronous condition. Multiple experiments are conducted, and the sum of the mean and three standard deviations of the linear fitting slope is taken to determine the second fault slope. Finally, the mean of the first fault-free slope and the second fault-free slope is calculated as the asynchronous slope threshold.

[0082] By processing the terminal to determine whether the linear fit goodness and linear fit slope match the cylinder asynchrony results, it can be determined whether there is a situation of asynchronous lifting of the left and right cylinders in the cylinder lifting structure. If so, calibration is required, thereby improving the calibration accuracy of the ceramic sheet detection device.

[0083] Step S4021: If so, calibrate the warpage detection device according to the warpage compensation height and the stroke fitting straight line.

[0084] If the processing terminal determines that the left and right cylinders in the cylinder lifting structure are not lifting synchronously, calibration is required. Therefore, the warpage detection device is calibrated based on the warpage compensation height and stroke fitting linearity. Specific calibration steps are detailed below. Figure 5 The steps in the process.

[0085] Step S4022: If not, obtain the no-load detection result.

[0086] If the processing terminal determines that there is no asynchronous lifting of the left and right cylinders in the cylinder lifting structure, it indicates that the abnormal lifting of the lifting block is caused by wear or debris interference of the lifting block itself. Therefore, the no-load test results are obtained to provide data support for the subsequent calibration of the warpage detection device.

[0087] The no-load detection result refers to the laser point cloud detection data on the upper surface of the lifting block. The processing terminal controls the picking robot arm to remove the ceramic piece at the current detection position when it is determined that there is no asynchronous lifting of the cylinders on the left and right sides, so that the lifting block is in a no-load state. The three-dimensional point cloud detection device is then activated to detect the upper surface of the lifting block, and the point cloud data is extracted and determined according to the coordinate range of the lifting block.

[0088] Step S4023: Calibrate the warpage detection device based on the no-load test results.

[0089] After determining the no-load test results, the warpage detection device is calibrated based on these results. Specific analysis steps are detailed below. Figure 6 The steps in the process.

[0090] Step S403: Obtain the deviation lifting block number.

[0091] Among them, the deviation lifting block number refers to the number of the lifting block that has an abnormal lifting situation and whose compensation height difference is greater than the upper limit threshold of the deviation. It is determined by the processing terminal by retrieving the number of the lifting block whose compensation height difference is greater than the upper limit threshold of the deviation during the current judgment process.

[0092] Step S404: Determine the warpage of the ceramic sheet based on the deviation lifting block number.

[0093] After determining the number of the deviation lifting block, the warpage of the ceramic sheet is determined by analyzing the number of the deviation lifting block. The specific analysis steps are as follows: Figure 8 The steps in the process.

[0094] The warpage of the ceramic sheet is consistent with that in step S1022. The processing terminal determines the warpage by numbering the lifting block according to the deviation. For specific methods and steps, please refer to [link / reference needed]. Figure 8 The steps in the process.

[0095] Reference Figure 5 The steps for calibrating the warp detection device based on the warp compensation height and stroke fitting straight line include: Step S500: Obtain the coordinates of the left cylinder and the right cylinder.

[0096] The coordinates of the left cylinder refer to the plane coordinates of the center point of the left lifting cylinder in the warping detection device under the point cloud detection coordinate system. The processing terminal calculates and determines the coordinates of the left cylinder based on the relative position of the center point of the left lifting cylinder with respect to the placement position of the ceramic piece under inspection and the plane coordinates of the ceramic piece under inspection under the point cloud detection coordinate system.

[0097] The coordinates of the right cylinder refer to the plane coordinates of the center point of the right lifting cylinder in the warping detection device under the point cloud detection coordinate system. The processing terminal calculates and determines the coordinates of the right cylinder based on the relative position of the center point of the right lifting cylinder with respect to the placement position of the ceramic piece under inspection and the plane coordinates of the ceramic piece under inspection under the point cloud detection coordinate system.

[0098] Step S501: Input the coordinates of the left cylinder and the right cylinder into the stroke fitting line to determine the cylinder stroke deviation.

[0099] Among them, cylinder stroke deviation refers to the stroke deviation between the left and right cylinders. It is determined by the processing terminal by first substituting the coordinates of the left cylinder and the right cylinder into the stroke fitting line to calculate the height corresponding to the coordinate point of the left cylinder and the coordinate point of the right cylinder, respectively, and then calculating the difference between the thrust height of the left cylinder and the thrust height of the right cylinder.

[0100] Step S502: Calculate the average value of the warp compensation height to determine the average compensation contact value.

[0101] The average value of the compensation contact refers to the average value of the warp compensation height, which is determined by the processing terminal by calculating the arithmetic mean of the warp compensation height.

[0102] Step S503: Input the cylinder stroke deviation, the average compensation contact value, and the preset reference stroke speed into the preset speed compensation model to determine the stroke compensation speed.

[0103] The reference stroke speed refers to the reference thrust speed of the cylinder initially set in the cylinder lifting structure. It is determined by the operator through offline lifting pre-experimentation of the ceramic plate, which ensures that the ceramic plate is lifted stably and will not be damaged during the lifting process. This is the reference stroke speed.

[0104] The speed compensation model is a formulaic model that calculates the stroke compensation speed based on cylinder stroke deviation, average compensation contact value, and reference stroke speed. The specific model formula is as follows: ; .

[0105] In the formula, To compensate for travel speed, This is due to cylinder stroke deviation. To compensate for the average contact value, This is the baseline travel speed.

[0106] Based on the premise of synchronous start and uniform lifting of the left and right servo cylinders, the speed compensation amount is derived. Under uniform motion, the stroke of the cylinder is the product of speed and time. The stroke deviation of the left and right cylinders is the difference in lifting height between the left and right cylinders due to the asynchronous accumulation of strokes, which eliminates the influence of the warping of the ceramic plate itself. This satisfies the product of speed deviation compensation amount and time. The average value of the compensation height is the actual lifting stroke of the cylinder at the reference speed during the current lifting process. Since the two times are the same, the first formula is derived and determined. Then, the first formula is transformed to determine the speed compensation amount that needs to be compensated and synchronizes the lifting speed of the left and right cylinders to the reference speed, which is the stroke compensation speed.

[0107] The stroke compensation speed refers to the total cylinder speed to be compensated corresponding to the stroke deviation of the left and right cylinders. It is calculated and determined by the processing terminal by inputting the cylinder stroke deviation, the average compensation contact value and the reference stroke speed into the speed compensation model.

[0108] Step S504: Calibrate the warpage detection device according to the stroke compensation speed.

[0109] After determining the stroke compensation speed, the processing terminal first calculates half of the stroke compensation speed. The speed of the cylinder on the side with a higher stroke is reduced by half of the stroke compensation speed, and the speed of the cylinder on the side with a lower stroke is increased by half of the stroke compensation speed. Constant lifting speeds for the left and right cylinders are generated and stored in the servo drive unit. After determining the compensation speed, the cylinder is controlled to lift again, and the compensation height is analyzed. If there is no abnormality in the lifting block, the speed adjustment amount is fixed. If there is still an abnormality in the lifting block, it is determined that there is a mid-stage jam or sudden change in mechanical resistance during the cylinder lifting process. The speed calibration process is then performed again for verification. If there is still an abnormality, a corresponding alarm prompt is given, and fault-related information is reported, so that the operator can maintain the cylinder.

[0110] Reference Figure 6 The steps for calibrating the warpage detection device based on the no-load test results include: Step S600: Perform linear fitting on the no-load test results to determine the top surface reference equation.

[0111] Among them, the top surface reference equation refers to the reference plane obtained by fitting the unloaded detection point cloud of each lifting block, which is determined by the processing terminal by performing least squares linear fitting on the unloaded detection results.

[0112] Step S601: Determine the maximum and minimum fitting height differences in the top surface reference equation based on the no-load test results.

[0113] The maximum fitting height difference refers to the maximum height difference between the detection height of each plane point in the no-load detection results and the fitting height of each plane point. The processing terminal first substitutes the xy plane coordinates of each detection point in the no-load detection results into the top surface reference equation to determine the fitting height of each detection point, then calculates the difference between the detection height and the fitting height of each point, and finally iterates through the differences to determine the maximum value of each difference, which is the maximum fitting height difference.

[0114] The minimum fitting height difference refers to the minimum height difference between the detection height of each plane point in the no-load detection results and the fitting height of each plane point. The processing terminal first substitutes the xy plane coordinates of each detection point in the no-load detection results into the top surface reference equation to determine the fitting height of each detection point, then calculates the difference between the detection height and the fitting height of each point, and finally iterates through the differences to determine the minimum value among all the differences, which is the minimum fitting height difference.

[0115] Step S602: Calculate the difference between the maximum fitted height difference and the minimum fitted height difference to determine the flatness of the lifting block.

[0116] Among them, the flatness of the lifting block refers to the quantitative index of the flatness of the upper surface of the lifting block, which is determined by the processing terminal by calculating the difference between the maximum and minimum fitted height differences.

[0117] Step S603: Calibrate the warpage detection device based on the flatness of the lifting block and the maximum fitting height difference.

[0118] After determining the flatness of the lifting block and the maximum fitted height difference, the warpage detection device is calibrated based on these parameters. Specific calibration steps are detailed below. Figure 7 The steps in the process.

[0119] Reference Figure 7 The steps for calibrating the warpage detection device based on the flatness of the lifting block and the maximum fitted height difference include: Step S700: Calculate the quotient of the maximum fitted height difference and the flatness of the lifting block to determine the lifting block protrusion coefficient.

[0120] Among them, the lifting block protrusion coefficient refers to the surface morphology characteristic judgment index of the lifting block surface caused by the unevenness defects, which is determined by the processing terminal by calculating the quotient of the maximum fitting height difference and the flatness of the lifting block.

[0121] Step S701: Determine whether the jacking block protrusion coefficient, maximum fitting height difference, and jacking block flatness meet the preset requirements for local debris results.

[0122] Among them, the local debris result refers to the lifting block protrusion coefficient being greater than the preset cleaning protrusion threshold, the maximum fitting height difference being greater than the preset cleaning height difference threshold, and the lifting block flatness being less than the preset cleaning flatness threshold. The requirement for the local debris result is that it is consistent with the local debris result.

[0123] The cleaning protrusion threshold refers to the lower limit of the jacking block protrusion coefficient when there is a risk of local debris and cleaning of the jacking block is required. If the jacking block protrusion coefficient is greater than the cleaning protrusion threshold, it indicates a risk of local debris. This threshold is determined by the operator through offline pre-experiments. The specific experimental steps are as follows: The operator conducts a jacking pre-experiment on a non-warped ceramic sheet under normal jacking conditions. Multiple jacking pre-experiments are conducted based on the smallest debris condition observed in historical operating conditions. The jacking block protrusion coefficient corresponding to the minimum debris condition when the jacking block is tested under no-load conditions is determined. The mean and standard deviation of the jacking block protrusion coefficient corresponding to the minimum debris condition are calculated. The sum of the mean and three times the standard deviation is calculated to determine the first protrusion coefficient. Multiple jacking pre-experiments are then conducted under debris-free conditions. The sum of the mean and three times the standard deviation of the jacking block protrusion coefficient is again calculated to determine the second protrusion coefficient. Finally, the average of the first and second protrusion coefficients is calculated, which is the cleaning protrusion threshold.

[0124] The cleaning height difference threshold refers to the lower limit of the maximum fitted height difference when there is a risk of local debris and cleaning of the lifting block is required. If the bulge coefficient of the lifting block is greater than the cleaning height difference threshold, it indicates a risk of local debris. This threshold is determined by the operator through offline pre-experiments. The specific experimental steps are as follows: The operator conducts a lifting pre-experiment on the non-warped ceramic sheet under normal lifting conditions. Multiple lifting pre-experiments are conducted based on the smallest debris situation observed in historical operating conditions to determine the maximum fitted height difference corresponding to the no-load test of the lifting block under the minimum debris condition. The mean and standard deviation of the maximum fitted height difference corresponding to the minimum debris condition are calculated. The sum of the mean and three times the standard deviation is calculated to determine the first height difference. Multiple lifting pre-experiments are then conducted under debris-free conditions to determine the sum of the mean and three times the standard deviation of the maximum fitted height difference again to determine the second height difference. Finally, the average of the first and second height differences is calculated, which is the cleaning height difference threshold.

[0125] The clean flatness threshold poses a risk of localized debris. Therefore, the upper limit of the flatness threshold for the lifting block needs to be determined when cleaning the lifting block. If the flatness of the lifting block exceeds the clean flatness threshold, it indicates a risk of localized debris. This is determined by the operator through offline pre-tests. The specific experimental steps are as follows: The operator conducts a lifting pre-test on a non-warped ceramic sheet under normal lifting conditions. Multiple lifting pre-tests are conducted using the smallest debris condition observed in historical operating conditions to determine the flatness of the lifting block under the minimum debris condition during no-load testing. The mean and standard deviation of the flatness of the lifting block corresponding to the minimum debris condition are calculated. The sum of the mean and three times the standard deviation is calculated to determine the first flatness. Multiple lifting pre-tests are then conducted under debris-free conditions to again determine the sum of the mean and three times the standard deviation of the lifting block flatness to determine the second flatness. Finally, the average of the first and second flatness is calculated, which is the clean flatness threshold.

[0126] By processing the terminal to determine whether the bulge coefficient, maximum fitting height difference, and flatness of the lifting block meet the requirements of the local debris results, it can be determined whether there are abnormal bulges on the lifting block, whether there is debris adhesion, and whether the lifting block needs to be cleaned. Based on the judgment results, the lifting block of the warpage detection device is cleaned to improve the calibration accuracy of the warpage detection device.

[0127] Step S7011: If yes, then start the preset cleaning device to clean and calibrate the warpage detection device.

[0128] If the processing terminal determines that the convexity coefficient, maximum fitting height difference, and flatness of the lifting block meet the requirements of the local debris results, it indicates that there is an abnormal convexity on the lifting block, debris is attached, and the lifting block needs to be cleaned. Therefore, the preset cleaning device is activated to clean the warpage detection device. After cleaning, the warpage detection device is re-inspected under no-load conditions. If the re-inspection results show that there is no abnormal convexity on the lifting block, no debris is attached, and the lifting block needs to be cleaned, and the deviation between the point cloud fitting plane of all lifting blocks and the reference no-load plane of the lifting block under the standard no-abnormal lifting condition is within the calibrated error range, then the cleaning of the warpage detection device is deemed complete, and the detection operation continues. Otherwise, the operator is alerted according to the maintenance alarm information calibrated in the system, and the corresponding unqualified lifting block number is given.

[0129] The cleaning device is a device used to clean debris and foreign objects from the lifting block. The cleaning device is installed on the warpage detection device and includes a robotic arm and a vacuum adsorption device. It is installed on the side of the detection station. When the lifting block needs to be cleaned, the processing terminal controls the robotic arm to move according to the number of the lifting block with abnormal protrusion that needs to be cleaned. The vacuum adsorption device is moved above the lifting block to be cleaned, adsorbs and cleans the foreign objects, and transfers the foreign objects to the dust collection bin through a hose. The negative pressure and cleaning time used by the cleaning device are determined by the operator based on the extraction pre-experimentation of common foreign objects on the lifting block.

[0130] Step S7012: If not, replace and calibrate the warpage detection device.

[0131] If the processing terminal determines that the jacking block protrusion coefficient, maximum fitting height difference, and jacking block flatness do not meet the requirements of the local debris results, it indicates that the warpage detection device cannot be calibrated by cleaning. Therefore, the warpage detection device is replaced and calibrated. An alarm is issued to the operator based on the maintenance alarm information calibrated in the system, and the corresponding unqualified jacking block number is given.

[0132] Reference Figure 8 The steps for determining the warpage of the ceramic sheet based on the deviation lifting block number include: Step S800: Obtain historical warp measurements.

[0133] Among them, historical warp detection refers to the warp detection results of ceramic pieces that flow to the next process within the historical valid judgment period. It is determined by the processing terminal by first extracting the detection warp of ceramic pieces within the valid judgment period and those that flow to subsequent processes.

[0134] Step S801: Perform numerical analysis on the historical warp values ​​to determine the historical warp mean and historical warp variance.

[0135] Among them, the historical warp mean refers to the average value of the warp detected in history, which is determined by the processing terminal by calculating the arithmetic mean of the warp detected in history.

[0136] Historical warp variance refers to the variance of historical warp, which is determined by the processing terminal by calculating the variance of historical warp.

[0137] Step S802: Obtain the warpage of the deviation station based on the deviation lifting block number and the preset flow waiting time.

[0138] The flow waiting time refers to the normal flow time for each ceramic piece to move to the next process after completing the inspection, excluding the re-inspection time when the lifting block is abnormal. That is, when the ceramic pieces are flowed using the sliding flow buffer mechanism, the time each ceramic piece stays in the current warp detection process. Under the sliding flow buffer mechanism, the ceramic pieces that have completed the inspection first enter the ceramic piece buffer area. The buffer area is dynamically updated as the inspection process and buffer flow time change. When the system determines that there is an abnormality in the warp during the forward inspection process, the ceramic pieces corresponding to the abnormal station in the sliding area are re-inspected, thereby improving the detection accuracy and efficiency of ceramic piece warp. The operator sets up an offline pre-experiment for fault detection. Based on the warp detection process, the mean and variance of the ceramic piece warp data in the buffer area are statistically significant, which can reflect the minimum number of ceramic piece samples corresponding to the warp abnormality and the determination of the ceramic piece detection speed during normal inspection.

[0139] The warpage of the deviation station refers to the warpage detection result of the station corresponding to the deviation lifting block number during the flow waiting time. It is determined by the processing terminal by first identifying the detection station where the deviation occurred according to the deviation lifting block number, and then retrieving the ceramic tile warpage detection data of the deviation detection station during the flow waiting time, with the current time as the end point and the flow waiting time as the length of time.

[0140] Step S803: Perform numerical analysis on the warpage of the deviation station to determine the mean and variance of the deviation warpage.

[0141] Among them, the mean value of deviation warpage refers to the mean value of the warpage of the deviation station, which is determined by the processing terminal by calculating the arithmetic mean of the warpage of the deviation station.

[0142] Deviation warpage variance refers to the variance of the warpage of the deviation station, which is determined by the processing terminal by calculating the variance of the warpage of the deviation station.

[0143] Step S804: Calculate the absolute deviation of the historical warp mean and the deviation warp mean to determine the warp mean deviation.

[0144] Among them, the warping mean deviation refers to the absolute deviation between the historical warping mean and the deviation warping mean, which is determined by the processing terminal by calculating the absolute deviation between the historical warping mean and the deviation warping mean.

[0145] Step S805: Calculate the absolute deviation of the historical warp variance and the deviation warp variance to determine the warp variance deviation.

[0146] Among them, the warping variance deviation refers to the absolute deviation in calculating the historical warping variance and the deviation warping variance, which is determined by the processing terminal by calculating the absolute deviation of the historical warping variance and the deviation warping variance.

[0147] Step S806: Determine whether the mean warpage deviation and the variance warpage deviation meet the preset requirements for normal flow results.

[0148] In this context, a normal workflow result refers to a warpage mean deviation that is less than the preset re-inspection mean difference threshold and a warpage variance that is less than the preset re-inspection variance difference threshold. A normal workflow result must be consistent with the normal workflow result.

[0149] The re-inspection mean difference threshold refers to the upper limit of the warping mean deviation when the mean of the ceramic pieces in the buffer area is not abnormal and re-inspection is not required. When the warping mean deviation of the ceramic pieces in the buffer area exceeds the re-inspection mean difference threshold, it indicates that the ceramic pieces in the buffer area need to be re-inspected. This threshold is determined by the operator through offline pre-experimentation. The specific experimental steps are as follows: The operator continuously conducts multiple batches of flow lifting warping detection experiments under the standard working conditions of normal ceramic pieces to be inspected, with no wear on the lifting block and no jamming in the cylinder synchronization. The maximum normal fluctuation boundary of the warping mean deviation when there is no equipment abnormality is determined according to the 3σ method. Then, a working condition is set where there are small debris on the lifting block and slight asynchrony of the cylinder, resulting in a slight abnormality in the warping offset of the workpiece. The minimum excess value of the warping mean deviation under the abnormal working condition is determined according to the 3σ method. The average of the maximum normal fluctuation boundary under the normal working condition and the minimum excess value under the abnormal working condition is taken as the re-inspection mean difference threshold.

[0150] The re-inspection variance threshold refers to the upper limit of the warpage variance deviation when the variance of the ceramic sheet in the buffer area is not abnormal and re-inspection is not required. When the warpage variance deviation of the ceramic sheet in the buffer area exceeds the re-inspection variance threshold, it indicates that the ceramic sheet in the buffer area needs to be re-inspected. This threshold is determined by the operator through offline pre-experimentation. The specific experimental steps are as follows: The operator continuously conducts multiple batches of flow lifting warpage detection experiments under the standard working conditions of normal ceramic sheet to be inspected, with no wear on the lifting block and no jamming in the cylinder synchronization. The maximum normal fluctuation boundary of the warpage variance deviation is determined according to the 3σ method when there is no equipment abnormality. Then, a working condition is set where there are small debris on the lifting block and slight asynchrony of the cylinder, resulting in a slight abnormality in the warpage offset of the workpiece. The minimum excess value of the warpage variance deviation under the abnormal working condition is determined according to the 3σ method. The average of the maximum normal fluctuation boundary under the normal working condition and the minimum excess value under the abnormal working condition is taken as the re-inspection variance threshold.

[0151] By processing the terminal to determine whether the mean deviation and variance deviation of warpage meet the requirements of normal flow results, it can be determined whether there is any abnormality in warpage during the pre-detection process. This eliminates the need for re-inspection of the ceramic sheet in the buffer area, thereby improving the detection accuracy of ceramic sheet warpage.

[0152] Step S8061: If so, re-inspect the preset ceramic sheet to determine the warpage of the ceramic sheet.

[0153] If the mean warpage deviation and variance warpage deviation are determined by the processing terminal to meet the requirements of normal flow results, it can be determined that there is no abnormal warpage during the pre-detection process. Therefore, it is not necessary to re-inspect the ceramic pieces in the buffer area. After the lifting block calibration is completed, the ceramic pieces currently being inspected are re-inspected to determine the warpage of the ceramic pieces.

[0154] Ceramic substrates refer to ceramic substrates that are tested by warpage detection devices. Ceramic substrates have a thin, planar structure and are used in fields such as electronic packaging, semiconductor substrates, and sensors. When the warpage of a ceramic substrate does not meet the standards, it will affect the operation of subsequent surface mount packaging processes and affect the function of the device. Therefore, it is necessary to test the warpage of ceramic substrates to eliminate those that do not meet the warpage standards.

[0155] The ceramic sheet warpage is consistent with the ceramic sheet warpage in step S404, and is the warpage of the ceramic sheet currently being inspected. When the processing terminal determines that there is no warpage abnormality in the pre-inspection process and there is no need to re-inspect the ceramic sheets in the buffer area, it re-inspects the ceramic sheet currently being inspected through the calibrated warpage detection device to determine the three-dimensional laser point cloud data of the ceramic sheet, and then calculates and determines the warpage of the ceramic sheet using the formula in step S1023.

[0156] Step S8062: If not, perform a retrospective inspection of the ceramic sheet to determine the warpage of the ceramic sheet.

[0157] If the processing terminal determines that there is an abnormality in the warpage during the pre-detection process, the ceramic pieces in the buffer area need to be re-inspected. In this case, the ceramic pieces are backtracked and re-inspected to determine the warpage of the ceramic pieces, thereby improving the detection accuracy of the warpage of the ceramic pieces.

[0158] The ceramic sheet warpage is consistent with the ceramic sheet warpage in step S404. It is the warpage of the currently inspected ceramic sheet and the ceramic sheet in the buffer area detected by the abnormal lifting block. When the processing terminal determines that there is an abnormal warpage during the pre-inspection process and the ceramic sheet in the buffer area needs to be re-inspected, the warpage of the currently inspected ceramic sheet and the ceramic sheet in the buffer area detected by the abnormal lifting block is re-inspected by the calibrated warpage detection device to determine the three-dimensional laser point cloud data of the ceramic sheet. Then, the ceramic sheet warpage is calculated and determined by the formula in step S1023.

[0159] Based on the same inventive concept, embodiments of this application provide a ceramic substrate warpage detection system based on laser three-dimensional point clouds, comprising: The acquisition module is used to acquire ceramic sheet contact height, point cloud detection data, warpage compensation height, lifting block coordinate range, no-load detection results, deviation lifting block number, left cylinder coordinates, right cylinder coordinates, historical detection warpage, and deviation station warpage. The memory is used to store the program for a method of detecting the warpage of ceramic substrates based on laser three-dimensional point clouds; The processor and memory can load and execute programs to implement a method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds.

[0160] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0161] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds, characterized in that, include: Acquire ceramic sheet contact height and point cloud detection data; Analyze the contact height of the ceramic sheet and the point cloud detection data to determine the abnormal results of the lifting block; Determine whether the abnormal result of the lifting block meets the preset requirements for abnormal lifting results; If so, obtain the warp compensation height; The warpage compensation height is analyzed to calibrate the preset warpage detection device and determine the warpage of the ceramic sheet; If not, perform data analysis on the point cloud detection data to determine the warpage of the ceramic sheet.

2. The method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds according to claim 1, characterized in that, The steps for analyzing ceramic sheet contact height and point cloud detection data to determine abnormal results of the lifting block include: The contact height of the ceramic sheet and the point cloud detection data were analyzed to determine the compensation contact height; Calculate the absolute deviation between the compensated contact height and the preset unbiased contact height to determine the compensation height difference; Determine whether the compensation height difference is greater than the preset upper limit threshold for deviation; If it is greater than the preset abnormal result, then the abnormal result will be determined as the abnormal result of the lifting block; If the result is not greater than the preset no-abnormal result, then the lifting block abnormal result will be determined.

3. The method for detecting the warpage of a ceramic substrate based on laser three-dimensional point clouds according to claim 2, characterized in that, The steps for analyzing ceramic tile contact height and point cloud detection data to determine the compensation contact height include: Plane fitting is performed on the point cloud detection data to determine the plane fitting equation; Obtain the coordinate range of the lifting block; Data extraction is performed on the point cloud detection data based on the coordinate range of the lifting block to determine the contact point cloud of the lifting block; Substitute the contact point cloud of the lifting block into the plane fitting equation to determine the reference fitting height; Calculate the height difference between the contact point cloud of the lifting block and the reference fitted height to determine the directional distance of the point cloud; Numerical analysis of the directed distance in the point cloud is performed to determine the directed distance quantiles; Calculate the difference between the ceramic tile contact height and the directional distance quantile to determine the compensation contact height.

4. The method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds according to claim 1, characterized in that, The steps for analyzing the warpage compensation height, calibrating the preset warpage detection device, and determining the warpage of the ceramic sheet include: Linear fitting is performed on the warpage compensation height to determine the goodness of linear fit and the stroke-fitted straight line; Data is extracted from the stroke-fitted straight line to determine the slope of the linear fit; Determine whether the goodness of linear fit and the slope of linear fit meet the preset requirements for cylinder asynchrony results; If so, the warpage detection device is calibrated based on the warpage compensation height and the straight line fitted by the stroke. If not, obtain the no-load test result; The warpage detection device is calibrated based on the no-load test results; Obtain the deviation lifting block number; The warpage of the ceramic sheet is determined based on the deviation lifting block number.

5. The method for detecting the warpage of a ceramic substrate based on laser three-dimensional point clouds according to claim 4, characterized in that, The steps for calibrating the warpage detection device based on the warpage compensation height and stroke fitting straight line include: Obtain the coordinates of the left and right cylinders; Input the coordinates of the left and right cylinders into the stroke fitting line to determine the cylinder stroke deviation; Calculate the average warpage compensation height to determine the average compensation contact value; The cylinder stroke deviation, the average compensation contact value, and the preset reference stroke speed are input into the preset speed compensation model to determine the stroke compensation speed. The warpage detection device is calibrated based on the travel compensation speed.

6. The method for detecting the warpage of a ceramic substrate based on laser three-dimensional point clouds according to claim 4, characterized in that, The steps for calibrating the warpage detection device based on the no-load test results include: Linear fitting is performed on the unloaded test results to determine the top surface reference equation; Based on the no-load test results, determine the maximum and minimum fitted height differences in the top surface reference equation; Calculate the difference between the maximum and minimum fitted height differences to determine the flatness of the lifting block; The warpage detection device is calibrated based on the flatness of the lifting block and the difference between the maximum fitted height.

7. The method for detecting the warpage of a ceramic substrate based on laser three-dimensional point clouds according to claim 6, characterized in that, The steps for calibrating the warpage detection device based on the difference between the flatness of the lifting block and the maximum fitted height include: Calculate the quotient of the maximum fitted height difference and the flatness of the lifting block to determine the lifting block protrusion coefficient; Determine whether the jacking block protrusion coefficient, maximum fitting height difference, and jacking block flatness meet the preset requirements for local debris results; If so, the preset cleaning device will be activated to clean and calibrate the warpage detection device; If not, the warpage detection device should be replaced and calibrated.

8. The method for detecting the warpage of a ceramic substrate based on laser three-dimensional point clouds according to claim 4, characterized in that, The steps for determining the warpage of the ceramic sheet based on the deviation lifting block number include: Obtain historical warp measurements; Numerical analysis was performed on historical warpage to determine the historical warpage mean and historical warpage variance; The warpage of the deviation station is obtained based on the deviation lifting block number and the preset flow waiting time; Numerical analysis was performed on the warpage of the deviation workstation to determine the mean and variance of the deviation warpage. Calculate the absolute deviation of the historical warp mean and the deviation warp mean to determine the warp mean deviation; Calculate the absolute deviation of the historical warp variance and the bias warp variance to determine the warp variance bias; Determine whether the mean warpage deviation and the variance warpage deviation meet the preset requirements for normal flow results; If so, the preset ceramic sheet will be re-inspected to determine the degree of warpage of the ceramic sheet; If not, the ceramic sheet will be retrospectively inspected to determine its warpage.

9. A ceramic substrate warpage detection system based on laser three-dimensional point cloud, characterized in that, include: The acquisition module is used to acquire the contact height of the ceramic sheet, point cloud detection data, and warpage compensation height. A memory for storing a program for detecting the warpage of a ceramic substrate based on laser three-dimensional point cloud as described in any one of claims 1 to 8; The processor and the program in the memory can be loaded and executed by the processor to implement the method for detecting the warpage of ceramic substrates based on laser three-dimensional point clouds as described in any one of claims 1 to 8.

Images