Control system and method of wafer box loading port

Abstract

The invention belongs to the technical field of intelligent control, and provides a control system and method for a wafer box loading port, and the method comprises the steps: obtaining a correlation adjustment coefficient and a suction cup adjustment number if the correlation between a suction connection area and a droop distance is tight, analyzing whether a suction cup distribution strategy in multiple historical taking processes is reasonable or not, and if the suction cup distribution strategy is not reasonable, judging whether the suction cup distribution strategy is reasonable or not; if yes, the distribution of the suction cups is adjusted according to the adjustment number of the suction cups, and a suction cup distribution diagram is constructed, so that on one hand, the probability that the thin wafer inclines in the taking process of the edge suction cups is reduced by adjusting the distance between the suction cups and the center of the thin wafer, and on the other hand, the probability that the thin wafer inclines is reduced by adjusting the circumferential angle of the suction cups on the thin wafer. Therefore, the circumferential angle distribution of the suction cup on the thin wafer is uniform, the influence on the stability degree of the suction cup in the thin wafer taking process is avoided, and the probability that the suction cup collides with the adjacent thin wafer in the thin wafer taking process is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control technology, and in particular relates to a control system and method for a wafer box loading port. Background Art

[0002] In semiconductor manufacturing, wafers, as core raw materials, are crucial for the quality of the final product. The precision and stability of their processing plays a decisive role. Cassette load ports, as a critical link in wafer transportation and storage, are crucial for the safe and efficient transfer of wafers between equipment. However, in actual production, cassette load ports face numerous technical challenges, severely impacting wafer processing efficiency and quality.

[0003] In existing technologies, it is difficult for workers to clearly understand the distribution spacing of thin wafers in three-dimensional space, making it difficult to formulate targeted handling strategies when subsequently picking up thin wafers. Furthermore, existing technologies cannot identify in real time the risk of instantaneous sagging caused by insufficient adsorption contact area. When the adsorption contact area is insufficient, the wafer may experience instantaneous sagging during the handling process. Therefore, during the wafer handling process, the instantaneous sagging distance of each historical handling process is compared with the corresponding thin wafer spacing for risk analysis, thereby identifying the instantaneous sagging risk in real time and ensuring the safety of the wafer handling process. Secondly, the distribution strategy of the suction cups on the thin wafer directly affects the stability of wafer picking. In the existing technology, there is a lack of effective analysis methods for whether the suction cup distribution strategy is reasonable. On the one hand, if the distance between the suction cup and the center of the thin wafer is not adjusted properly, the edge suction cup will easily cause the thin wafer to tilt during the picking process, increasing the risk of wafer damage. On the other hand, the uneven distribution of the suction cups on the thin wafer will affect the stability of the suction cups during the picking process, and it is also easy for the wafer to collide with adjacent thin wafers during the picking process. Therefore, the suction cup distribution strategy in multiple historical picking processes is analyzed to see whether it is reasonable. If it is unreasonable, the suction cup distribution is adjusted according to the number of suction cup adjustments. The technology of constructing a suction cup distribution map can improve the stability of wafer picking and reduce the probability of collision.

[0004] To this end, the present invention provides a control system and method for a wafer box loading port. Summary of the Invention

[0005] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0006] The technical solution adopted by the present invention to solve its technical problem is: In a first aspect, a method for controlling a wafer cassette load port includes: Based on the length, width and height of the wafer box, a 3D model of the wafer box is constructed. The thin wafer area within the 3D model is extracted, and the distribution spacing of the thin wafers is analyzed to obtain the thin wafer spacing. Obtain the instantaneous sag distance during each historical pick-up process, compare it with the corresponding inter-crystalline spacing, and determine the comparison risk level. If the risk level is high, perform a correlation analysis based on the adsorption contact area during each historical pick-up process to determine the correlation between the adsorption area and the sag distance. If the suction area and the droop distance are closely correlated, the correlation adjustment coefficient is obtained, the number of suction cup adjustments is calculated, and the rationality of the suction cup distribution strategy in the historical picking process is analyzed. If it is unreasonable, the distribution of the suction cups will be adjusted according to the number of suction cups, and a suction cup distribution map will be constructed.

[0007] Preferably, the process of obtaining the intergranular spacing is as follows: A 3D model of the wafer box is constructed using the length of the wafer box as the Y-axis, the width of the wafer box as the X-axis, and the height of the wafer box as the Z-axis. The thin wafer area within the 3D model of the wafer box is extracted, and adjacent thin wafers within the thin wafer area are combined to obtain multiple groups of thin wafer combinations. The distance between adjacent thin crystals in each thin crystal combination is obtained as the thin crystal spacing.

[0008] Preferably, the process of determining the comparison risk is as follows: Obtain the actual Z-axis coordinate of the thin wafer on the cassette 3D model after each historical take-up, and subtract the absolute value from the theoretical Z-axis coordinate of the thin wafer on the cassette 3D model, and calculate the ratio with the Z-axis length to obtain the instantaneous sag distance; The ratio of the instantaneous sag distance to the corresponding thin crystal spacing is calculated and the comparison risk is output.

[0009] Preferably, if the risk is high, a correlation analysis is performed based on the adsorption contact area during each historical pickup process. The process is as follows: If the comparison risk is greater than the comparison risk threshold, a high-risk signal is displayed. Taking the shape of the bottom of the suction cup as an example, the circle is used to calculate the area of ​​the bottom of each suction cup. The sum is calculated and the ratio is calculated with the surface area of ​​the thin wafer to obtain the adsorption contact area. The adsorption contact area and comparison risk in each historical picking process are used as a set of association analysis data groups to obtain multiple sets of association analysis data groups. The adsorption contact area and comparison risk in each set of association analysis data groups are extracted respectively, and sorted and summarized according to the time series corresponding to the historical picking process to obtain the adsorption area sequence and comparison risk sequence; In the comparison risk sequence, adjacent comparison risk degrees are combined to obtain adjacent risk comparison combinations. The adjacent adsorption contact areas in each group of adjacent risk comparison combinations are subtracted, and the adjacent comparison risk degrees are subtracted. The absolute values ​​are taken to obtain the single group adsorption difference and the single group risk difference.

[0010] Preferably, the correlation between the suction area and the sag distance is determined as follows: The Pearson correlation coefficient calculation method was used to process the data of all single-group adsorption differences and single-group risk differences to obtain the correlation analysis value; If the correlation analysis value is greater than or equal to the correlation analysis threshold, it is displayed as a close correlation signal.

[0011] Preferably, the correlation adjustment coefficient is obtained to obtain the suction cup adjustment quantity, and the process is as follows: All unit correlation coefficients are averaged and calculated, and the correlation adjustment coefficient is output; The ratio of the comparison risk threshold and the associated adjustment coefficient is calculated, and the adjusted suction area is output. The adjusted suction area is summed with the current suction area to obtain the adjusted suction area, and the ratio is calculated with the adsorption area corresponding to each suction cup to obtain the suction cup adjustment quantity.

[0012] Preferably, the suction cup distribution strategy in the historical picking process is analyzed to see whether it is reasonable. The process is as follows: On the surface of the thin wafer, the center of the bottom circle of each suction cup and the center of the thin wafer surface are extracted, and the distance between the center of the bottom circle of each suction cup and the center of the thin wafer surface is obtained. The ratio of the distance to the circumference of the thin wafer surface is calculated as the unit radial distance. All unit radial distances are averaged and the output is the radial distance mean; Connect the centers of the bottom circles of adjacent suction cups with straight lines, and connect the centers of the bottom circles of adjacent suction cups with the centers of the thin wafer surfaces with straight lines to construct a fitting triangle; Get the distance between the centers of the bottom circles of adjacent suction cups as the side length of the fitted triangle A; The distances between the centers of the bottom circles of adjacent suction cups and the centers of the thin wafer surfaces are used as the side lengths of the fitted triangles B and C, respectively. Input the side lengths of the fitted triangle A, B, and C into the cosine theorem formula, calculate the ratio with the circumferential angle, and output the unit circumferential angle; The circumferential angle difference and the radial distance mean are summed and the output is the distribution strategy evaluation value. If the distribution strategy evaluation value is greater than the distribution strategy evaluation threshold, it is displayed as a distribution strategy low match signal.

[0013] Preferably, the distribution of the suction cups is adjusted according to the number of suction cups, and the process is as follows: Calculate the ratio of the circumferential angle to the number of suction cup adjustments and output the average circumferential angle; The distribution strategy evaluation threshold is calculated by ratio with the number of suction cup adjustments, and the adjusted radial value is output. The adjusted radial value is then subtracted from the unit radial distance to obtain the unit radial adjustment amount.

[0014] Preferably, the process of constructing the suction cup distribution map is as follows: Based on the adjusted radial value and the average circumferential angle, each suction cup is placed on the corresponding angle line on the thin wafer surface to construct a suction cup distribution map.

[0015] In a second aspect, a control system for a wafer cassette load port includes: Constructing an extraction module: Based on the length, width, and height of the wafer box, a 3D model of the wafer box is constructed. The thin wafer area within the 3D model is extracted, and the distribution spacing of the thin wafers is analyzed to obtain the thin wafer spacing. Correlation analysis module: Obtain the instantaneous sag distance during each historical pick-up process, compare it with the corresponding thin crystal spacing, and determine the comparison risk level. If the risk level is high, perform correlation analysis based on the adsorption contact area during each historical pick-up process to determine the correlation between the adsorption area and the sag distance; Adjustment Evaluation Module: If there is a close correlation between the suction area and the droop distance, the correlation adjustment coefficient is obtained, the suction cup adjustment quantity is calculated, and the suction cup distribution strategy in the historical picking process is analyzed to determine whether it is reasonable. Distribution adjustment module: If it is unreasonable, the distribution of the suction cups will be adjusted according to the number of suction cups to construct a suction cup distribution map.

[0016] The beneficial effects of the present invention are as follows: The present invention constructs a three-dimensional model of a wafer cassette based on the wafer placement environment within the wafer cassette, extracts the thin wafer region, obtains the distribution spacing of thin wafers within the thin wafer region, and obtains the thin-crystal spacing. This facilitates analysis of the adjacent relationship of thin wafers in multiple spatial directions, allowing personnel to more clearly understand the distribution spacing of thin wafers in three-dimensional space, formulate targeted picking strategies for subsequent thin wafer picking, and reduce the probability of collision with adjacent thin wafers during the picking process. By obtaining the instantaneous sag distance during each historical picking process and performing a risk comparison analysis with the corresponding thin-crystal spacing, a comparison risk degree is determined. If the risk is high, a correlation analysis is performed based on the suction contact area during each historical picking process to determine the correlation between the suction area and the sag distance. This helps to identify the instantaneous sag risk caused by insufficient suction contact area in real time, triggering the loading port control system to automatically adjust the suction cup parameters to avoid collision with the wafer vertically below. In addition, combined with the analysis of the uniform distribution of thin wafers in the thin wafer region, basic data for design optimization is provided for the thin wafer loading density (spacing) within the wafer cassette, enabling targeted optimization of the thin wafer loading density within the wafer cassette. According to the present invention, if the correlation between the suction area and the droop distance is relatively close, the correlation adjustment coefficient is obtained, the number of suction cup adjustments is obtained, and whether the suction cup distribution strategy in multiple historical picking processes is reasonable is analyzed. If it is unreasonable, the distribution of the suction cups is adjusted according to the number of suction cup adjustments, and a suction cup distribution map is constructed. On the one hand, by adjusting the distance between the suction cup and the center of the thin wafer, the probability of the edge suction cup tilting during the picking process is reduced. On the other hand, by adjusting the circumferential angle of the suction cup on the thin wafer, the circumferential angle of the suction cup on the thin wafer is evenly distributed, thereby avoiding affecting the stability of the suction cup during the thin wafer picking process and reducing the probability of collision with adjacent thin wafers during the thin wafer picking process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 It is a flowchart of the steps of a control method of a wafer box loading port of the present invention; Figure 2 It is a judgment flow chart of a control method of a wafer box loading port of the present invention; Figure 3 It is a schematic diagram of a control system of a wafer box loading port of the present invention. DETAILED DESCRIPTION

[0019] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0020] Example 1:

[0021] like Figure 1 - Figure 2 As shown, a method for controlling a wafer box loading port according to an embodiment of the present invention includes: Step 1: Perform spatial transformation on the wafer placement environment in the wafer box, construct a 3D model of the wafer box, extract the thin wafer area, and obtain the distribution spacing of the thin wafers in the thin wafer area to obtain the thin wafer spacing; In some embodiments, a 3D model of the wafer box is constructed with the length of the interior of the wafer box as the Y axis, the width of the interior of the wafer box as the X axis, and the height of the interior of the wafer box as the Z axis. The specific process is as follows: For example, the top camera and four side cameras are calibrated separately using a checkerboard method to determine the focal length, focus position, and distortion coefficient of each camera. After calibration, each pixel coordinate acquired by each camera is converted into a physical coordinate, and a three-dimensional model of the crystal box is constructed based on the converted physical coordinates. Extracting a thin wafer region within the wafer box three-dimensional model, and combining adjacent thin wafers within the thin wafer region to obtain multiple thin wafer combinations; It should be noted that the plurality of thin crystal combinations include horizontally adjacent thin crystals and vertically adjacent thin crystals, wherein the horizontally adjacent thin crystals include X-axis horizontally adjacent thin crystals and Y-axis horizontally adjacent thin crystals, and the vertically adjacent thin crystals include Z-axis vertically adjacent thin crystals; Obtain the distance between adjacent thin crystals in each thin crystal combination and perform average calculation to obtain the thin crystal spacing; The method for obtaining the distance between adjacent thin crystals in each thin crystal combination is to use each thin crystal in the thin wafer area as a thin crystal point in the three-dimensional model of the crystal box, use the coordinate distance formula to obtain the distance between adjacent thin crystal points in each thin crystal combination, and calculate the ratio of the distance to the length of the corresponding coordinate axis in the three-dimensional model of the crystal box; For example, if the thin crystal combination being analyzed is adjacent thin crystals horizontally along the X axis or adjacent thin crystals horizontally along the Y axis, then after using the coordinate distance formula to obtain the distance between adjacent thin crystal points in each thin crystal combination, the ratio of this distance to the X or Y coordinate axis length (the width or length inside the wafer box) in the corresponding wafer box three-dimensional model is calculated. The thin crystal combination analyzed is adjacent thin crystals perpendicular to the Z axis. After using the coordinate distance formula to obtain the distance between adjacent thin crystal points in each thin crystal combination, the ratio of this distance to the Z coordinate axis length (the height inside the wafer box) in the corresponding wafer box 3D model is calculated. Specifically, obtaining the inter-crystalline spacing is beneficial for: combining adjacent thin wafers within a thin wafer region into multiple thin crystal groups, including horizontally adjacent thin crystals (horizontally adjacent thin crystals along the X axis and the Y axis) and vertically adjacent thin crystals (vertically adjacent thin crystals along the Z axis). This allows for analysis of the adjacent relationships of thin wafers from multiple spatial directions, enabling staff to more clearly understand the distribution spacing of thin wafers in three-dimensional space; Secondly, due to the limited space within the wafer cassette, thin wafers are prone to collisions with adjacent thin wafers during retrieval. Therefore, after understanding the distribution spacing of thin wafers in three-dimensional space, a targeted retrieval strategy can be developed for subsequent thin wafer retrieval, reducing the probability of collisions with adjacent thin wafers during retrieval. Step 2: During multiple historical thin wafer removal processes, obtain the instantaneous sag distance during each historical removal process and perform a risk comparison analysis with the corresponding thin wafer spacing to determine the comparison risk level. If the risk is high, perform a correlation analysis based on the adsorption contact area during each historical removal process to determine the correlation between the adsorption area and the sag distance. In some embodiments, the process for determining the comparison risk is as follows: The laser displacement sensor is used to obtain the actual Z-axis coordinate of the thin wafer on the 3D model of the wafer box after each historical pick-up. The actual Z-axis coordinate of the thin wafer on the 3D model of the wafer box is subtracted from the theoretical Z-axis coordinate of the thin wafer on the 3D model of the wafer box, and the absolute value is calculated by ratio calculation with the Z-axis length to obtain the instantaneous sag distance. Calculate the ratio of the instantaneous sag distance to the corresponding inter-crystalline spacing and output the comparison risk; It should be noted that the corresponding inter-crystalline spacing means: for example, when taking a thin wafer from a wafer box, due to the limited space inside the wafer box, it can only be taken out by horizontal extraction. However, when the suction cup adsorbs the thin wafer and takes it out, it will sag instantly, which can easily cause it to collide with the thin wafer placed vertically below. Therefore, the corresponding inter-crystalline spacing is the distance between the thin wafer being taken out and the thin wafer placed vertically below it when taking the thin wafer out; If the comparison risk is less than or equal to the comparison risk threshold, it means that during the analyzed historical handling process, the instantaneous sagging phenomenon when the thin wafer is sucked by the suction cup, resulting in a collision with the vertically placed thin wafer below, is less likely; If the comparison risk is greater than the comparison risk threshold, it means that during the historical handling process being analyzed, the instantaneous sagging phenomenon when the thin wafer was sucked by the suction cup caused a high risk of collision with the vertically placed thin wafer below. If the risk probability is high, the adsorption contact area and the comparison risk degree are correlated and analyzed as follows: Taking the shape of the bottom of the suction cup as a circle, the area of ​​the bottom of each suction cup is obtained as the adsorption contact area using the circle area calculation formula; It should be noted that during each historical picking process, each suction cup was in complete contact with the surface of the thin wafer, and there was no incomplete contact between the suction cup and the surface of the thin wafer, that is, there was a gap between the suction cup and the surface of the thin wafer; The adsorption area corresponding to each suction cup is summed and the ratio is calculated with the surface area of ​​the thin wafer to obtain the adsorption contact area; It should be noted that the surface of a thin wafer is usually circular, and the area can also be obtained using the circular area calculation formula; The adsorption contact area and the comparison risk in each historical picking process are used as a set of correlation analysis data groups to obtain multiple sets of correlation analysis data groups; Extract the adsorption contact area in each group of correlation analysis data, and sort and summarize it according to the corresponding time series of historical extraction to obtain the adsorption area sequence; In the adsorption area sequence, adjacent adsorption contact areas are combined to obtain adjacent adsorption combinations; Similarly, extract the comparison risk within each group of association analysis data, and sort and summarize the corresponding time series according to history to obtain the comparison risk sequence; In the comparison risk sequence, adjacent comparison risk degrees are combined to obtain adjacent comparison risk combinations; In adjacent adsorption combinations, the adjacent adsorption contact areas are subtracted and the absolute value is taken to obtain the single group adsorption difference; In adjacent risk comparison combinations, the adjacent risk comparisons are subtracted and the absolute value is taken to obtain the risk difference of a single group; The Pearson correlation coefficient calculation method was used to process the data of all single-group adsorption differences and single-group risk differences. The specific process is as follows: S1, respectively obtain all single group adsorption differences and the average values ​​corresponding to all single group risk differences, and obtain the mean adsorption difference mean risk difference ; S2, respectively obtain the covariance corresponding to all single-group adsorption differences and the covariance corresponding to all single-group relative risk differences, obtain the adsorption covariance and relative risk covariance, and perform sum calculation to obtain the numerator in the Pearson correlation coefficient calculation formula m represents the total number of single-group adsorption differences, which is also the total number of single-group relative risk differences, where the total number of single-group adsorption differences is equal to the total number of single-group relative risk differences; S3, respectively obtain the sum of squares of deviations corresponding to all single-group adsorption differences and the sum of squares of deviations corresponding to all single-group relative risk differences, and obtain the sum of squares of adsorption deviations and relative risk deviations as the denominator in the Pearson correlation coefficient calculation formula m represents the total number of single-group adsorption differences, which is also the total number of single-group relative risk differences, where the total number of single-group adsorption differences is equal to the total number of single-group relative risk differences; S4, according to the numerator and denominator in the Pearson correlation coefficient calculation formula, input it into the improved Pearson correlation coefficient calculation formula, and calculate and process to obtain the correlation analysis value ; Specifically, ; If the correlation analysis value is greater than or equal to the correlation analysis threshold, it means that the correlation between the adsorption contact area and the comparison risk is relatively close, which is displayed as a close correlation signal; If the correlation analysis value is less than the correlation analysis threshold, it means that the correlation between the adsorption contact area and the comparison risk is not close, which is displayed as a non-close correlation signal; It should be noted that after using the Pearson correlation coefficient calculation formula to calculate the data of all single-group adsorption differences and single-group risk differences, the calculated values ​​are in In this range, the original Pearson correlation coefficient calculation formula is now processed with absolute value, so the calculated value is , because the calculation principle of the Pearson correlation coefficient formula is that the closer it is to 1, the stronger the linear positive correlation strength is, the closer it is to -1, the stronger the linear negative correlation strength is, and the closer it is to 0, the weaker the linear correlation strength is; Specifically, correlation analysis is beneficial in: It can identify in real time the risk of instantaneous drooping caused by insufficient suction contact area, triggering the load port control system to automatically adjust the suction cup parameters to avoid collision with the wafer below; Secondly, to increase the suction contact area, the bottom of the suction cup can be designed with a curved suction piece and a sealing groove can be added. This provides an adjustment idea for the suction cup bottom design and reduces the risk of wafer slippage due to unstable suction. Furthermore, combined with the analysis of the uniform distribution of thin wafers within the thin wafer area, basic data for design optimization of the thin wafer loading density (spacing) within the wafer cassette is provided, enabling targeted optimization of the thin wafer loading density within the wafer cassette. The solution of this embodiment is as follows: a three-dimensional model of the wafer box is constructed based on the wafer placement environment in the wafer box, and the thin wafer area is extracted. The distribution of thin wafers in the thin wafer area is evenly analyzed to diagnose whether the thin wafer distribution is uniform. This is conducive to analyzing the adjacent relationship of thin wafers in multiple spatial directions, so that the staff can more clearly understand the distribution spacing of thin wafers in three-dimensional space, formulate targeted picking strategies for subsequent thin wafer picking, and reduce the probability of collision with adjacent thin wafers during the picking process. By obtaining the instantaneous sag distance during each historical picking process and comparing it with the corresponding thin wafer spacing, the staff can obtain the instantaneous sag distance during each historical picking process and compare it with the corresponding thin wafer spacing. Conduct risk comparison analysis to determine the degree of risk. If the risk is relatively high, conduct a correlation analysis based on the adsorption contact area during each historical pickup process to determine the correlation between the adsorption area and the droop distance. This helps to identify the instantaneous droop risk caused by insufficient adsorption contact area in real time, triggering the loading port control system to automatically adjust the suction cup parameters to avoid collision with the wafer vertically below. In addition, combined with the uniform distribution analysis of thin wafers in the thin wafer area, it provides basic data for design optimization of the thin wafer loading density in the wafer box, and optimizes the design of the thin wafer loading density in the wafer box in a targeted manner.

[0022] Example 2:

[0023] like Figure 1 - Figure 2 As shown, based on Example 1, a control method for a wafer box loading port according to an embodiment of the present invention further includes: Step 3: If the suction area and the droop distance are closely correlated, obtain the correlation adjustment coefficient and the number of suction cup adjustments. The suction cup distribution strategy from multiple historical pick-up processes is then used to obtain a distribution strategy evaluation value to assess whether the suction cup distribution strategy is reasonable. In some embodiments, the adsorption contact area and the comparison risk in each group of association analysis data are calculated as a ratio, and the unit association coefficient is output; The unit correlation coefficients of all correlation analysis data groups are averaged and the correlation adjustment coefficients are output; Calculate the ratio of the comparison risk threshold and the correlation adjustment coefficient, output the adjusted adsorption area, and sum it with the current adsorption area to output the adjusted adsorption area; Calculate the ratio of the adjusted suction area to the suction area corresponding to each suction cup, and output the suction cup adjustment quantity; Adjust the number of suction cups and perform a uniform distribution analysis based on the current suction cup adsorption strategy. The process is as follows: A1: On the surface of the thin wafer, extract the center of the bottom circle of each suction cup and the center of the thin wafer surface. Obtain the distance between the center of each suction cup and the center of the thin wafer surface, and calculate the ratio of this distance to the circumference of the thin wafer surface as the unit radial distance. Average all unit radial distances and output the radial distance mean. It should be noted that there is a torque balance directly related to the wafer's droop risk and stress. Specifically, if the contact point between the suction cup and the thin wafer deviates too far from the wafer's center, an eccentric torque around the wafer's center can easily form, causing the wafer to tilt during handling. Therefore, obtaining the unit radial distance is helpful in determining the potential tilt risk of the thin wafer during handling from a two-dimensional plane dimension. A2: Connect the centers of the bottom circles of adjacent suction cups with straight lines, and connect the centers of the bottom circles of adjacent suction cups with the centers of the thin wafer surfaces with straight lines to construct a fitting triangle. Get the distance between the centers of the bottom circles of adjacent suction cups as the side length of the fitted triangle A; The distances between the centers of the bottom circles of adjacent suction cups and the centers of the thin wafer surfaces are used as the side lengths of the fitted triangles B and C, respectively. Input the length of the side of the fitted triangle A, the length of the side of the fitted triangle B and the length of the side of the fitted triangle C into the cosine theorem formula and compare them with the circumferential angle ( ) to calculate the ratio and output the unit circumferential angle ; Specifically, the triangle function formula is: ,in, Expressed as the side length of the fitted triangle A, Expressed as the side length of the fitted triangle B, and It is expressed as the length of the side of the fitted triangle C; Based on the arrangement of the suction cup bottoms on the thin wafer surface, adjacent suction cup bottoms are combined in a clockwise direction to obtain a suction cup arrangement combination, wherein the suction cup arrangement combination also includes the suction cup bottom arranged first and the suction cup bottom arranged last; Input the unit circumferential angles corresponding to the suction cup arrangement and combination into the Euclidean distance formula in a clockwise manner, and output the circumferential angle difference; The circumferential angle difference and the radial distance mean are summed and the output is the distribution strategy evaluation value; It can be understood that the meaning of the distribution strategy evaluation value is: by combining the radial distance mean and the circumferential angle difference, the rationality of the current suction cup distribution strategy is quantitatively evaluated. On the one hand, the radial distance mean reflects the distance between the center of the bottom of each suction cup and the center of the surface of the thin wafer. Since the overall distribution of the suction cups deviates far from the center of the thin wafer, it is very easy to form an eccentric torque around the center of the thin wafer. Moreover, in actual operation, if the suction cups are mainly distributed in the edge area of ​​the thin wafer, and there are fewer suction cups in the central area, then when picking up the thin wafer, the pulling force of the edge suction cups on the thin wafer will cause the thin wafer to have a rotation tendency around the center, thereby tilting. On the other hand, the circumferential angle difference reflects the distribution angle of adjacent suction cups in the circumferential direction of the thin wafer. Since the suction cups are unevenly distributed in the circumferential direction, the thin wafer is subjected to uneven circumferential force during the picking process, which in turn affects the stability of the picking and causes the thin wafer to rotate or twist during the picking process. If the distribution strategy evaluation value is greater than the distribution strategy evaluation threshold, it means that the suction cup distribution strategy in multiple historical picking processes has a low match with the current adjusted number of suction cups, and there is a high potential tilt risk, which is displayed as a low distribution strategy match signal; If the distribution strategy evaluation value is less than or equal to the distribution strategy evaluation threshold, it means that the suction cup distribution strategy in multiple historical picking processes has a high degree of match with the current adjusted number of suction cups, and there is a low potential tilt risk, which is displayed as a high distribution strategy match signal; Step 4: If the suction cup distribution strategy does not match the suction cup adjustment quantity, adjust the suction cup distribution based on the suction cup adjustment quantity to construct a suction cup distribution map. In some embodiments, the ratio of the circumferential angle to the number of suction cup adjustments is calculated and the average circumferential angle is output; Based on the average circumferential angle, each suction cup is placed on the corresponding angle line on the thin wafer surface, and the radial distance of each suction cup is adjusted. The process is as follows: It should be noted that the angle line is based on the center of the thin wafer surface as the starting point, the average circumferential angle as the reference angle, and a straight line is drawn from the starting point to the edge of the thin wafer surface. Each angle line corresponds to only one suction cup. Calculate the ratio of the distribution strategy evaluation threshold to the suction cup adjustment number, output the adjusted radial value, and subtract it from the unit radial distance to obtain the unit radial adjustment amount; Based on the adjusted radial value and the average circumferential angle, each suction cup is placed on the corresponding angle line on the thin wafer surface to construct a suction cup distribution map; The solution of this embodiment is: if the correlation between the suction area and the droop distance is relatively close, then the correlation adjustment coefficient is obtained, the number of suction cup adjustments is obtained, and whether the suction cup distribution strategy in multiple historical picking processes is reasonable is analyzed. If it is unreasonable, the distribution of the suction cups is adjusted according to the number of suction cup adjustments, and a suction cup distribution map is constructed. On the one hand, by adjusting the distance between the suction cup and the center of the thin wafer, the probability of the edge suction cup tilting during the picking process is reduced. On the other hand, by adjusting the circumferential angle of the suction cup on the thin wafer, the circumferential angle of the suction cup on the thin wafer is evenly distributed, avoiding affecting the stability of the suction cup during the thin wafer picking process and reducing the probability of collision with adjacent thin wafers during the thin wafer picking process.

[0024] Example 3:

[0025] like Figure 3 As shown, a control system for a wafer box loading port according to an embodiment of the present invention includes: Constructing an extraction module: Based on the length, width, and height of the wafer box, a 3D model of the wafer box is constructed. The thin wafer area within the 3D model is extracted, and the distribution spacing of the thin wafers is analyzed to obtain the thin wafer spacing. Correlation analysis module: Obtain the instantaneous sag distance during each historical pick-up process, compare it with the corresponding thin crystal spacing, and determine the comparison risk level. If the risk level is high, perform correlation analysis based on the adsorption contact area during each historical pick-up process to determine the correlation between the adsorption area and the sag distance; Adjustment Evaluation Module: If there is a close correlation between the suction area and the droop distance, the correlation adjustment coefficient is obtained, the suction cup adjustment quantity is calculated, and the suction cup distribution strategy in the historical picking process is analyzed to determine whether it is reasonable. Distribution adjustment module: If it is unreasonable, the distribution of the suction cups will be adjusted according to the number of suction cups to construct a suction cup distribution map.

[0026] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling a wafer cassette loading port, characterized in that: include: Based on the length, width and height of the wafer box, a 3D model of the wafer box is constructed. The thin wafer area within the 3D model is extracted, and the distribution spacing of the thin wafers is analyzed to obtain the thin wafer spacing. Obtain the instantaneous sag distance during each historical pick-up process, compare it with the corresponding inter-crystalline spacing, and determine the comparison risk level. If the risk level is high, perform a correlation analysis based on the adsorption contact area during each historical pick-up process to determine the correlation between the adsorption area and the sag distance. If the suction area and the droop distance are closely correlated, the correlation adjustment coefficient is obtained, the number of suction cup adjustments is calculated, and the rationality of the suction cup distribution strategy in the historical picking process is analyzed. If it is unreasonable, the distribution of the suction cups will be adjusted according to the number of suction cups, and a suction cup distribution map will be constructed.

2. The method for controlling a wafer cassette loading port according to claim 1, wherein: The process of obtaining the thin crystal spacing is as follows: A 3D model of the wafer box is constructed using the length of the wafer box as the Y-axis, the width of the wafer box as the X-axis, and the height of the wafer box as the Z-axis. The thin wafer area within the 3D model of the wafer box is extracted, and adjacent thin wafers within the thin wafer area are combined to obtain multiple groups of thin wafer combinations. The distance between adjacent thin crystals in each thin crystal combination is obtained as the thin crystal spacing.

3. The method for controlling a wafer cassette loading port according to claim 1, wherein: The process of determining the comparison risk is as follows: Obtain the actual Z-axis coordinate of the thin wafer on the cassette 3D model after each historical take-up, and subtract the absolute value from the theoretical Z-axis coordinate of the thin wafer on the cassette 3D model, and calculate the ratio with the Z-axis length to obtain the instantaneous sag distance; The ratio of the instantaneous sag distance to the corresponding thin crystal spacing is calculated and the comparison risk is output.

4. The method for controlling a wafer cassette loading port according to claim 1, wherein: If the risk is high, a correlation analysis is performed based on the adsorption contact area during each historical pickup process. The process is as follows: If the comparison risk is greater than the comparison risk threshold, a high-risk signal is displayed. Taking the shape of the bottom of the suction cup as an example, the circle is used to calculate the area of ​​the bottom of each suction cup. The sum is calculated and the ratio is calculated with the surface area of ​​the thin wafer to obtain the adsorption contact area. The adsorption contact area and comparison risk in each historical picking process are used as a set of association analysis data groups to obtain multiple sets of association analysis data groups. The adsorption contact area and comparison risk in each set of association analysis data groups are extracted respectively, and sorted and summarized according to the time series corresponding to the historical picking process to obtain the adsorption area sequence and comparison risk sequence; In the comparison risk sequence, adjacent comparison risk degrees are combined to obtain adjacent risk comparison combinations. The adjacent adsorption contact areas in each group of adjacent risk comparison combinations are subtracted, and the adjacent comparison risk degrees are subtracted. The absolute values ​​are taken to obtain the single group adsorption difference and the single group risk difference.

5. The method for controlling a wafer cassette loading port according to claim 4, wherein: The correlation between the suction area and the sag distance is determined as follows: The Pearson correlation coefficient calculation method was used to process the data of all single-group adsorption differences and single-group risk differences to obtain the correlation analysis value; If the correlation analysis value is greater than or equal to the correlation analysis threshold, it is displayed as a close correlation signal.

6. The method for controlling a wafer cassette loading port according to claim 1, wherein: Get the associated adjustment coefficient and obtain the suction cup adjustment quantity. The process is as follows: All unit correlation coefficients are averaged and calculated, and the correlation adjustment coefficient is output; The ratio of the comparison risk threshold and the associated adjustment coefficient is calculated, and the adjusted suction area is output. The adjusted suction area is summed with the current suction area to obtain the adjusted suction area, and the ratio is calculated with the adsorption area corresponding to each suction cup to obtain the suction cup adjustment quantity.

7. The method for controlling a wafer cassette loading port according to claim 1, wherein: Analyze whether the suction cup distribution strategy in the historical picking process is reasonable. The process is as follows: On the surface of the thin wafer, the center of the bottom circle of each suction cup and the center of the thin wafer surface are extracted, and the distance between the center of the bottom circle of each suction cup and the center of the thin wafer surface is obtained. The ratio of the distance to the circumference of the thin wafer surface is calculated as the unit radial distance. All unit radial distances are averaged and the output is the radial distance mean; Connect the centers of the bottom circles of adjacent suction cups with straight lines, and connect the centers of the bottom circles of adjacent suction cups with the centers of the thin wafer surfaces with straight lines to construct a fitting triangle; Get the distance between the centers of the bottom circles of adjacent suction cups as the side length of the fitted triangle A; The distances between the centers of the bottom circles of adjacent suction cups and the centers of the thin wafer surfaces are used as the side lengths of the fitted triangles B and C, respectively. Input the side lengths of the fitted triangle A, B, and C into the cosine theorem formula, calculate the ratio with the circumferential angle, and output the unit circumferential angle; The circumferential angle difference and the radial distance mean are summed and the output is the distribution strategy evaluation value. If the distribution strategy evaluation value is greater than the distribution strategy evaluation threshold, it is displayed as a distribution strategy low match signal.

8. The method for controlling a wafer cassette loading port according to claim 1, wherein: Adjust the distribution of suction cups according to the number of suction cups. The process is as follows: Calculate the ratio of the circumferential angle to the number of suction cup adjustments and output the average circumferential angle; The distribution strategy evaluation threshold is calculated by ratio with the number of suction cup adjustments, and the adjusted radial value is output. The adjusted radial value is then subtracted from the unit radial distance to obtain the unit radial adjustment amount.

9. The method for controlling a wafer cassette loading port according to claim 8, wherein: The process of constructing the suction cup distribution map is as follows: Based on the adjusted radial value and the average circumferential angle, each suction cup is placed on the corresponding angle line on the thin wafer surface to construct a suction cup distribution map.

10. A control system for a wafer cassette load port, characterized in that: include: Constructing an extraction module: Based on the length, width, and height of the wafer box, a 3D model of the wafer box is constructed. The thin wafer area within the 3D model is extracted, and the distribution spacing of the thin wafers is analyzed to obtain the thin wafer spacing. Correlation analysis module: Obtain the instantaneous sag distance during each historical pick-up process, compare it with the corresponding thin crystal spacing, and determine the comparison risk level. If the risk level is high, perform correlation analysis based on the adsorption contact area during each historical pick-up process to determine the correlation between the adsorption area and the sag distance; Adjustment Evaluation Module: If there is a close correlation between the suction area and the droop distance, the correlation adjustment coefficient is obtained, the suction cup adjustment quantity is calculated, and the suction cup distribution strategy in the historical picking process is analyzed to determine whether it is reasonable. Distribution adjustment module: If it is unreasonable, the distribution of the suction cups will be adjusted according to the number of suction cups to construct a suction cup distribution map.

Images