A PCB separating path dynamic optimization method based on real-time stress feedback

Abstract

This invention discloses a dynamic optimization method for PCB depaneling paths based on real-time stress feedback, relating to the field of PCB manufacturing technology. The method includes: acquiring PCB panel structure data, initial depaneling path data, and stress detection location data; reading connection area location data from the PCB panel structure data and forming a connection area set; dividing a candidate depaneling path segment set based on the initial depaneling path data; and establishing a path connection table based on the positional correspondence between the cutting coverage of the candidate depaneling path segments and the connection area set; acquiring real-time stress feedback data and executed depaneling path data during the depaneling process; and removing cut connection areas from the connection area set according to the path connection table to form a remaining connection area set. This invention can identify cut connection areas, remaining connection areas, and connection areas to be cut based on the path connection table, establishing a stable data foundation for real-time stress feedback data and depaneling path control.

Description

Technical Field

[0001] This invention relates to the field of PCB processing technology, specifically to a dynamic optimization method for PCB depaneling paths based on real-time stress feedback. Background Technology

[0002] After PCB panelization is completed, multiple sub-boards are usually separated from the panelization structure through a separation process. During the separation process, the initial separation path is usually predetermined based on the PCB panelization structure data, sub-board outline boundaries, connection area position data, and equipment processing path. The separation execution mechanism cuts the connection areas sequentially according to the predetermined path, thereby completing the separation of the sub-boards. This method can meet the needs of continuous path execution and cutting position control in conventional separation operations. In the actual process of panel separation, there is usually a structural force transmission relationship between different connection areas. After a candidate panel separation path segment is executed, its cutting action may change the stress state of the remaining connection areas. When panel separation is executed sequentially based solely on the initial panel separation path data, there is still room for improvement in the linkage evaluation between real-time stress feedback data, the degree of influence of completed panel separation path segments, and the stress release risk of subsequent connection areas to be cut. Therefore, it is necessary to combine PCB panel structure data, initial board separation path data, stress detection location data, and real-time stress feedback data to establish a set of connection areas, a set of candidate board separation path segments, and a path connection table. During the board separation process, a set of remaining connection areas, regional stress records, cutting contribution, virtual transfer amount, and board separation release risk value are generated. This allows the selection of subsequent board separation path segments to be dynamically adjusted based on the executed board separation path data and real-time stress status, thereby improving the stress release control capability during the board separation path execution process. Therefore, this invention proposes a dynamic optimization method for PCB board separation paths based on real-time stress feedback. Summary of the Invention

[0003] The purpose of this invention is to provide a dynamic optimization method for PCB depaneling paths based on real-time stress feedback, so as to solve the problems mentioned in the background art.

[0004] This invention can be achieved through the following technical solution: a dynamic optimization method for PCB depaneling paths based on real-time stress feedback, comprising: Step 1: Obtain PCB panel structure data, initial board separation path data, and stress detection location data. Read the connection area location data from the PCB panel structure data and form a connection area set. Divide the candidate board separation path segment set according to the initial board separation path data, and establish a path connection table based on the positional correspondence between the cutting coverage of the candidate board separation path segments and the connection area set. Step 2: During the plate separation process, obtain real-time stress feedback data and the data of the plate separation path that has been executed. Remove the cut connection areas from the connection area set according to the path connection table to form the remaining connection area set. Then, based on the stress detection location data and real-time stress feedback data, form the regional stress record of each remaining connection area. Step 3: For the completed plate-splitting path segment, calculate the difference in regional stress feedback value and the difference in regional stress change rate based on the regional stress records before and after cutting, and determine the path interval of the connection area according to the path connection table to form the cutting contribution. Step 4: Read the subsequent board splitting path segments that have not been written with executed board splitting path data from the candidate board splitting path segment set, temporarily remove the corresponding connection areas to be cut from the remaining connection area set to form a virtual connection area set, and form a virtual transfer amount based on the cutting contribution amount. Form the board splitting release risk value based on the virtual transfer amount and the regional stress record. Step 5: Determine the next execution path segment based on the risk value of each subsequent board segment release, forming an optimized board segment path, and control the board segment execution mechanism to execute the subsequent board segments according to the optimized board segment path.

[0005] A further technical improvement of the present invention is that, in step one, after forming a set of connection areas and establishing a path connection table, stress detection positions are written to the connection areas in the set of connection areas, including: Read the connection area location data from the PCB panel structure data, and determine the connection area boundary of each connection area based on the connection area location data; Based on the path connection table, read the candidate board path segment corresponding to each connection area, and determine the start coverage position and end coverage position based on the cut-off coverage range of the candidate board path segment. The cut-off extension baseline is formed by the start coverage position and end coverage position. The midpoint is determined by the extension length along the boundary of the connecting area in the direction of the cut-off extension baseline, and this midpoint is determined as the midpoint of the boundary of the connecting area; Read the clamping constraint area from the PCB panel structure data, and connect the midpoint of the connection area boundary with the boundary position of the clamping constraint area closest to the connection area to form a clamping force transmission baseline. With the midpoint of the boundary of the connecting area as the origin, the direction of the cutting extension baseline as the first coordinate direction, and the direction of the clamping force transmission baseline as the second coordinate direction, a local coordinate system for the connecting area is formed. Map the stress detection location in the stress detection location data to the local assigned coordinate system to obtain the cutting direction projection coordinates of the stress detection location in the first coordinate direction and the clamping direction projection coordinates in the second coordinate direction. Based on the path segment order adjacency relationship of the candidate sub-board path segments corresponding to the connection area in the path connection table and the position adjacency relationship of the connection area boundary, the adjacent connection areas are determined. For adjacent connection regions, the boundary sampling position between two adjacent connection regions is taken as the position to be determined. Each position to be determined is mapped to the local home coordinate system corresponding to the two adjacent connection regions. The projection coordinates of the position to be determined relative to the two adjacent connection regions in the cutting direction and the clamping direction are obtained. The projection distance value of the position to be determined relative to the two adjacent connection regions is formed based on the projection coordinates of the cutting direction and the clamping direction. The home boundary line is formed based on the two positions to be determined with the same projection distance value. The stress detection domain of the connection zone is formed by the boundary lines surrounding the same connection zone. The stress detection locations that fall within the stress detection domain are written into the data fields of the corresponding connection area, forming a set of connection areas with stress detection locations written in them.

[0006] A further technical improvement of the present invention is that: in step one, the candidate board path segment set is divided according to the initial board path data, including: The boundary of each connection area is determined based on the connection area location data, and the path extension direction of the initial board splitting path and the cutting coverage range of each connection area covered by the initial board splitting path are read based on the initial board splitting path data. For each connecting area, along the path extension direction of the initial partition path, the position where the cut-off coverage first enters the boundary of the connecting area is determined as the entry point to the side boundary, and the position where the cut-off coverage last leaves the boundary of the connecting area is determined as the exit point to the side boundary. The cutting coverage direction is formed based on the direction from the entry point to the exit point of the side boundary, and the cutting coverage length is formed based on the path length along the initial partition path between the entry point and the exit point of the side boundary. The entry point to the side boundary, the exit point from the side boundary, the cut-off coverage direction, and the cut-off coverage length are written into the corresponding connection area to form cut-off coverage pattern data; the cut-off coverage pattern data of each connection area are sorted according to the order in which the initial board path covers each connection area to form a cut-off coverage pattern sequence. Based on the entry and exit side boundary positions of each connection area, the cut-off coverage area corresponding to each connection area is divided into a candidate board path segment, so that each candidate board path segment corresponds to a connection area. For two adjacent connected regions in the cut-and-cover pattern sequence, read the path interval formed along the initial plate path between the departure side boundary position of the previous connected region and the entry side boundary position of the next connected region. When the path interval does not cover any connected regions other than the two adjacent connected regions, and the projection direction of the cut-and-cover direction of the previous connected region and the next connected region is consistent in the path extension direction, a head-and-tail release connection relationship is formed. Calculate the difference between the cut-off coverage length of the subsequent connection zone and the cut-off coverage length of the previous connection zone, and form the cut-off coverage length difference direction based on the direction of change of the value of the difference; Based on the relationship between the beginning and end release and the direction of the difference in the cut-off coverage length, the path segment order adjacency relationship between adjacent candidate partition path segments is determined; A set of candidate board path segments is formed according to the order in which the initial board path covers the connecting area and the sequential adjacency relationship between adjacent candidate board path segments.

[0007] A further technical improvement of the present invention lies in: in step two, forming regional stress records for each remaining connection zone based on stress detection location data and real-time stress feedback data, including: Before executing the current candidate board segment, read the execution start time of the candidate board segment, and collect the real-time stress feedback data of the stress detection positions written in each remaining connection area before the execution start time, arranged in adjacent order of sampling time, to form a pre-cut-off sampling window; After the candidate board segment being cut off is completed, the execution termination time of the candidate board segment is read, and the real-time stress feedback data of the stress detection positions written in each remaining connection area are collected after the execution termination time and arranged in adjacent order of sampling time to form a sampling window after cutting off. For the same remaining connection area, the real-time stress feedback data corresponding to each stress detection position at the same sampling time are averaged according to the number of stress detection positions to form the regional stress feedback value at that sampling time. The regional stress change is formed based on the difference in regional stress feedback values ​​between adjacent sampling times, and the regional stress change rate is formed based on the time interval between the regional stress change and adjacent sampling times. The regional stress feedback value, regional stress change amount, regional stress change rate, and sampling time are written into the corresponding remaining connection area to form a regional stress record.

[0008] A further technical improvement of the present invention is that, in step three, forming the cutting contribution includes: For each completed board segment and each remaining connection area, the difference in regional stress feedback values ​​between adjacent sampling times is read from the regional stress record before the cut of the completed board segment, and the average of the absolute values ​​of the difference in regional stress feedback values ​​between adjacent sampling times is determined as the reference disturbance amount before the cut. Read the difference in stress feedback values ​​of the remaining connection area before and after the cut of the completed plate-splitting path segment. Subtract the reference disturbance before the cut from the absolute value of the difference in stress feedback values ​​before and after the cut. If the subtraction result is less than 0, the effective area stress increment value is determined as 0. If the subtraction result is greater than or equal to 0, the subtraction result is determined as the effective area stress increment value. The direction of the effective area stress increment is determined according to the direction of change of the difference in stress feedback values ​​before and after the cut. Read the difference in the rate of change of stress in the remaining connection area before and after the cut, and determine the direction of the difference in the rate of change of stress in the area based on the direction of the change of the numerical value of the difference in the rate of change of stress in the area before and after the cut. Based on whether the difference in the rate of change of stress in the region before and after the cut is 0, and whether the direction of the stress increment in the effective region is consistent with the direction of the difference in the rate of change of stress in the region, the combined calculation result of the value of the stress increment in the effective region and the absolute value of the difference in the rate of change of stress in the region before and after the cut is determined. When the effective area stress increment is 0, the shearing contribution is determined to be 0. When the effective area stress increment value is greater than 0 and the difference in the stress change rate before and after the cut is 0, the effective area stress increment value is determined as the combined calculation result, and the combined calculation result is divided by the value 1 plus the path interval of the connection area to form the cut contribution. When the effective stress increment value is greater than 0 and the direction of the effective stress increment is consistent with the direction of the difference in the rate of change of the stress in the region, the effective stress increment value is added to the absolute value of the difference in the rate of change of the stress in the region before and after the cut, forming the same-direction cut-off gain. The same-direction cut-off gain is determined as the combined calculation result. The combined calculation result is divided by the value 1 plus the path interval of the connection area to form the cut-off contribution. When the effective stress increment value is greater than 0 and the direction of the effective stress increment is opposite to the direction of the difference in the rate of change of regional stress, the absolute value of the difference between the effective stress increment value and the absolute value of the difference in the rate of change of regional stress before and after the cut is determined as the reverse cut offset amount. The reverse cut offset amount is determined as the combined calculation result. The combined calculation result is divided by the value 1 plus the path interval of the connection area to form the cut contribution amount.

[0009] A further technical improvement of the present invention is that: in step four, the method for forming a virtual transfer amount based on the cutoff contribution includes: After temporarily removing the connection area to be cut from the remaining connection area set and forming a virtual connection area set, the connection area boundary of the connection area to be cut before temporary removal is read as the original connection area boundary. Read virtual connection regions from the set of virtual connection regions that have boundary segments that overlap with the original connection region boundary, and form the transfer entry connection region; Taking the transfer entry connection area as the first path level, the candidate sub-board path segments corresponding to each virtual connection area are read according to the path connection table, and the path level of each virtual connection area in the virtual connection area set is determined according to the path segment order adjacency relationship between the candidate sub-board path segments. For each virtual connection region, the boundary adjacency length between the virtual connection region and each cut connection region is read and accumulated to form the cut boundary adjacency length. The boundary adjacency length between the virtual connection region and the original connection region boundary is read. The virtual boundary openness is formed by the ratio between the sum of the cut boundary adjacency length and the boundary adjacency length between the virtual connection region and the original connection region boundary and the connection region boundary length of the virtual connection region. The preset hierarchical correction relationship includes the correspondence between path levels and hierarchical correction values; Based on the path hierarchy, the corresponding hierarchy correction value is read from the preset hierarchy correction relationship, and the virtual transfer amount of the virtual connection area is formed based on the cutoff contribution, virtual boundary openness and hierarchy correction value.

[0010] A further technical improvement of the present invention is that: in step four, the method for forming a plate release risk value based on the virtual transfer amount and regional stress record includes: For each virtual connection region in the set of virtual connection regions, the regional stress feedback value corresponding to the sampling time adjacent to the temporary removal time of the subsequent plate path segment is read from the regional stress record and used as the current regional stress feedback value. The current regional stress feedback value is added to the virtual transfer amount corresponding to the virtual connection region to form the predicted regional stress value after virtual execution. Read the maximum and average predicted values ​​from the predicted values ​​of the virtual post-execution region corresponding to each virtual connection area, and form a prediction set difference based on the difference between the maximum and average predicted values; Read the virtual transfer amount corresponding to each virtual connection area. When the sum of all virtual transfer amounts is 0, the transfer concentration ratio is set to 0. When the sum of all virtual transfer amounts is greater than 0, the transfer concentration ratio is formed based on the ratio of the maximum virtual transfer amount to the sum of all virtual transfer amounts. Multiply the predicted concentration difference by the transfer concentration ratio to form the risk value of the board release.

[0011] A further technical improvement of the present invention is that, in step five, the next execution path segment is determined based on the risk value of the board release corresponding to each subsequent board path segment, including: A risk sorting sequence is formed based on the risk value of each subsequent board segment release, and the process of forming a risk sorting sequence and determining the next execution path segment is called the path selection process. The same subsequent board path segment that is within the preset number of consecutive preset times at the end of the risk sorting sequence is identified as the risk retention path segment. Subsequent board-segmented path segments that have at least one of the following relationships with the risk-delayed path segment corresponding to the area to be cut off: boundary adjacency relationship and path segment sequence adjacency relationship; are written into the set of retreat-associated path segments. For each yielding associated path segment, the corresponding connection area to be cut off is temporarily removed from the set of remaining connection areas. Then, the risk value of the risk retention path segment is re-formed according to the above-mentioned method for forming the risk value of the risk retention path segment. The risk yielding difference is formed based on the original risk value of the risk retention path segment and the re-formed risk value of the risk retention path segment. The path switching distance is formed based on the path length along the initial board path between the starting positions of the retreat-related path segment and the risk retention path segment, and the retreat release amount is formed based on the risk retreat difference and the path switching distance. The yield-related path segment that ranks first in yield release amount is determined as the next execution path segment.

[0012] Compared with the prior art, the present invention has the following beneficial effects: This invention forms a set of connection areas by reading connection area location data from PCB panel structure data, divides a set of candidate board separation path segments based on initial board separation path data, and then establishes a path connection table based on the positional correspondence between the cutting coverage of the candidate board separation path segments and the set of connection areas, thus forming a clear data association between the candidate board separation path segments and the connection areas. As a result, the cut connection areas, remaining connection areas, and connection areas to be cut can be identified based on the path connection table, establishing a stable data foundation for real-time stress feedback data and board separation path control. Furthermore, during the plate separation process, this invention generates regional stress records for each remaining connection zone based on real-time stress feedback data and stress detection location data. It also calculates the difference in regional stress feedback values, the difference in regional stress change rates, and the path interval of the connection zone for the completed plate separation path segments to form the cutting contribution. Thus, the influence of the completed plate separation path segments on the remaining connection zones is no longer inferred solely from the path sequence, but is jointly determined by the regional stress records before and after cutting and the path connection relationship, giving the cutting influence a clear data source and calculation basis. On the other hand, when selecting subsequent board splitting path segments, the present invention temporarily removes the connection areas to be cut from the remaining connection area set to form a virtual connection area set, and forms a virtual transfer amount based on the cutting contribution, and then forms a board splitting release risk value in combination with the regional stress record; then, the next execution path segment is determined according to the board splitting release risk value corresponding to each subsequent board splitting path segment to form an optimized board splitting path, and the board splitting execution mechanism is controlled to execute the subsequent board splitting according to the optimized board splitting path; thus, the board splitting path can be dynamically adjusted according to the real-time stress feedback results and the virtual release risk assessment results, reducing the board splitting risk caused by local stress concentration during the subsequent board splitting process. Attached Figure Description

[0013] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0014] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0015] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0016] Please see Figure 1 As shown, this invention provides a dynamic optimization method for PCB depaneling paths based on real-time stress feedback, including: Step 1: Obtain PCB panel structure data, initial board separation path data, and stress detection location data. Read the connection area location data from the PCB panel structure data and form a connection area set. Divide the candidate board separation path segment set according to the initial board separation path data, and establish a path connection table based on the positional correspondence between the cutting coverage of the candidate board separation path segments and the connection area set. Specifically, in this embodiment, PCB panel structure data, initial board separation path data, and stress detection location data are first acquired. The PCB panel structure data includes the panel outline boundary, sub-board outline boundary, connection area location data, and clamping constraint area location; the initial board separation path data includes the path coordinates, path extension direction, path start position, path end position, and cutting coverage area of ​​the initial board separation path; the stress detection location data includes the position coordinates of each stress detection location in the PCB panel coordinate system and the detection location number; after forming the stress detection domain corresponding to each connection area, there must be at least one stress detection location in the stress detection location data that can be positionally matched with that stress detection domain. The connection area location data is read from the PCB panel structure data, and the connection area boundary of each connection area is determined based on the connection area location data. The connection area boundary is formed by the outer perimeter boundary position of the corresponding physical connection part in the connection area location data. The boundary sampling position is read along the boundary extension direction of the connection area boundary. When the number of boundary sampling positions is odd, the middle boundary sampling position is determined as the middle position of the connection area boundary. When the number of boundary sampling positions is even, the position between the two middle boundary sampling positions is determined as the middle position of the connection area boundary, thus forming a set of connection areas.

[0017] After forming the set of connection regions, the path extension direction of the initial board-splitting path is read based on the initial board-splitting path data, and the cut-off coverage range of each connection region covered by the initial board-splitting path is also read. For each connection region, along the path extension direction of the initial board-splitting path, the position where the cut-off coverage range first enters the boundary of the connection region is determined as the entry side boundary position, and the position where the cut-off coverage range last leaves the boundary of the connection region is determined as the exit side boundary position. The cut-off coverage direction is formed based on the direction from the entry side boundary position to the exit side boundary position, and the cut-off coverage length is formed based on the path length along the initial board-splitting path from the entry side boundary position to the exit side boundary position. Subsequently, the entry side boundary position, exit side boundary position, cut-off coverage direction, and cut-off coverage length are written into the corresponding connection region to form cut-off coverage pattern data. The cut-off coverage pattern data of each connection region is sorted according to the order in which the initial board-splitting path covers each connection region to form a cut-off coverage pattern sequence.

[0018] For each connection area, based on the entry and exit side boundary positions of the connection area, the cut-off coverage range corresponding to the connection area is divided into a candidate board path segment. The connection area number, entry side boundary position, exit side boundary position, cut-off coverage direction, and cut-off coverage length of the connection area are written into the corresponding candidate board path segment, so that each candidate board path segment corresponds to a connection area.

[0019] For adjacent preceding and following connecting regions in the cut-and-cover pattern sequence, the path interval formed along the initial segmentation path between the departure side boundary of the preceding connecting region and the entry side boundary of the following connecting region is read. When this path interval does not cover any connecting regions other than the preceding and following connecting regions, and the projection directions of the cut-and-cover directions of the preceding and following connecting regions onto the path extension direction are consistent, a head-to-tail release connection relationship is formed between the candidate segmentation path segment corresponding to the preceding connecting region and the candidate segmentation path segment corresponding to the following connecting region.

[0020] After establishing the beginning and end release connection, calculate the difference between the cut-off coverage length of the subsequent connection zone and the cut-off coverage length of the preceding connection zone. When the cut-off coverage length difference is greater than 0, the direction of the cut-off coverage length difference is determined as the direction of length increase. When the cut-off coverage length difference is less than 0, the direction of the cut-off coverage length difference is determined as the direction of length decrease. When the cut-off coverage length difference is equal to 0, the direction of the cut-off coverage length difference is determined as the direction of no change in length.

[0021] Based on the beginning-end release connection relationship and the direction of the cut-off coverage length difference, the path segment order adjacency relationship between adjacent candidate board path segments is determined. Specifically, for the preceding and following connecting areas in the cut-off coverage pattern sequence, when a beginning-end release connection relationship is formed between the candidate board path segments corresponding to the preceding and following connecting areas, the direction of the cut-off coverage length difference corresponding to the ordered adjacent connecting areas is read; when the ordered adjacent connecting areas forming a beginning-end release connection relationship are read for the first time, the direction of the cut-off coverage length difference corresponding to the ordered adjacent connecting areas is determined as the current order adjacency reference direction, and the candidate board path segments corresponding to the preceding and following connecting areas corresponding to the ordered adjacent connecting areas are determined to have a path segment order adjacency relationship; subsequent ordered adjacent connecting areas... When a connection zone forms a head-to-tail release relationship, and the direction of the cut-off coverage length difference corresponding to the subsequent adjacent connection zone is consistent with the direction of the current adjacent reference position, the candidate board path segment corresponding to the previous connection zone and the candidate board path segment corresponding to the subsequent adjacent connection zone are determined to have a path segment sequential adjacent relationship. When an adjacent connection zone does not form a head-to-tail release relationship, or the direction of the cut-off coverage length difference corresponding to the adjacent connection zone is inconsistent with the direction of the current adjacent reference position, the candidate board path segment corresponding to the previous connection zone and the candidate board path segment corresponding to the subsequent connection zone are determined to not have a path segment sequential adjacent relationship.

[0022] The path segment sequence is written to each candidate board path segment according to the order in which the initial board path covers each connection area. The adjacency relationship of the path segment sequence is then written to the data field of adjacent candidate board path segments, forming a set of candidate board path segments. The set of candidate board path segments includes the path segment number, path start position, path end position, path segment sequence, cut-off coverage area, corresponding connection area number, and path segment sequence adjacency relationship of each candidate board path segment.

[0023] After forming a set of candidate board segment paths, a path connection table is established based on the positional correspondence between the cut-off coverage of the candidate board segment paths and the set of connection areas. For any candidate board segment path, when the cut-off coverage of the candidate board segment path covers the boundary of a connection area, the path segment number, path segment sequence, cut-off coverage, and connection area number of the candidate board segment path are written into the path connection table, thus establishing a positional correspondence between the candidate board segment path and the connection area. The path connection table is used to subsequently identify cut-off connection areas from the set of connection areas and to read the candidate board segment path corresponding to each connection area.

[0024] After forming a set of connection areas and establishing a path connection table, stress detection positions are written for the connection areas in the set. For each connection area, the candidate board segment corresponding to that connection area is read according to the path connection table, and the start and end coverage positions are determined based on the cutting coverage range of the candidate board segment. The start and end coverage positions form a cutting extension baseline. Then, the clamping constraint area is read from the PCB panel structure data, and the midpoint of the connection area boundary is connected to the boundary position of the clamping constraint area closest to the connection area to form a clamping force transmission baseline.

[0025] Using the midpoint of the connection zone boundary as the origin, the direction of the cutting extension baseline as the first coordinate direction, and the direction of the clamping force transmission baseline as the second coordinate direction, a local coordinate system for the connection zone is formed. The stress detection positions in the stress detection location data are mapped to this local coordinate system to obtain the projected coordinates of the stress detection position in the cutting direction (first coordinate direction) and the projected coordinates in the clamping direction (second coordinate direction). Subsequently, based on the path segment order adjacency relationship of the candidate plate path segments corresponding to the connection zone in the path connection table and the positional adjacency relationship of the connection zone boundaries, adjacent connection zones are determined.

[0026] For adjacent connection areas, the boundary sampling position between two adjacent connection areas is taken as the position to be determined. Each position to be determined is mapped to the local home coordinate system corresponding to the two adjacent connection areas to obtain the cutting direction projection coordinates and clamping direction projection coordinates of the position to be determined relative to the two adjacent connection areas. Based on the cutting direction projection coordinates and clamping direction projection coordinates, the absolute value of the cutting direction projection coordinates and the absolute value of the clamping direction projection coordinates are added to form the projection distance value of the position to be determined relative to the corresponding connection area. The projection distance difference between the projection distance values ​​of the position to be determined relative to the two adjacent connection areas is calculated. The projected distance difference is sequentially read from the boundary sampling positions between two adjacent connection areas. When a boundary sampling position with a projected distance difference of 0 exists, this position is determined as the location through which the boundary line passes. When no boundary sampling position with a projected distance difference of 0 exists, the two adjacent boundary sampling positions where the direction of change of the projected distance difference value changes is read, and the location through which the boundary line passes is determined between these two adjacent boundary sampling positions based on the ratio of the absolute values ​​of the projected distance differences corresponding to these two adjacent boundary sampling positions. When no two adjacent boundary sampling positions where the direction of change of the projected distance difference value changes is read, the boundary sampling position with the smallest absolute value of the projected distance difference is determined as the location through which the boundary line passes. The locations through which the boundary line passes are connected in the order of boundary extension to form the boundary line.

[0027] After the stress detection domain of a connection zone is formed by the boundary lines surrounding the same connection zone, the stress detection positions in the stress detection location data are matched with the stress detection domain. Stress detection positions falling within the stress detection domain are then written into the data field of the corresponding connection zone, forming a set of connection zones with written stress detection positions. This set of connection zones with written stress detection positions is used to determine the stress detection positions corresponding to each of the remaining connection zones after the remaining connection zone set is formed.

[0028] Step 2: During the plate separation process, obtain real-time stress feedback data and the data of the plate separation path that has been executed. Remove the cut connection areas from the connection area set according to the path connection table to form the remaining connection area set. Then, based on the stress detection location data and real-time stress feedback data, form the regional stress record of each remaining connection area. Specifically, after forming the connection area set, candidate board segmentation path segment set, and path connection table, the board segmentation process begins. When the executed board segmentation path data is empty and no cutting contribution has been formed, the candidate board segmentation path segment with the smallest path segment order in the candidate board segmentation path segment set is determined as the first executed path segment, and the board segmentation execution mechanism is controlled to execute this first executed path segment. After the first executed path segment completes the cutting and forms the corresponding area stress record, the subsequent cutting contribution formation, board segmentation release risk value formation, and path selection process begins. Real-time stress feedback data and executed board segmentation path data are acquired during the board segmentation process. Real-time stress feedback data is collected at fixed sampling time intervals. Executed board segmentation path data includes the path segment number of the completed board segmentation path segment, the execution start time, the execution end time, and the actual cutting coverage area. The connection area corresponding to the completed board segmentation path segment is read according to the path connection table, and this connection area is determined as the cut connection area; then, the cut connection area is removed from the connection area set to form the remaining connection area set. When the executed board splitting path data is empty, all connection areas in the connection area set are used as remaining connection areas; when at least one completed board splitting path segment is written into the executed board splitting path data, the corresponding disconnected connection areas are removed one by one according to the path connection table, so that the remaining connection area set is updated synchronously with the board splitting execution process.

[0029] After forming the set of remaining connection regions in the current board segmentation state, for the currently executing candidate board segmentation path segment, the execution start time of that candidate board segmentation path segment is read. Based on the data fields of each remaining connection region in the set of remaining connection regions, the stress detection positions written to that remaining connection region are read, and real-time stress feedback data of these stress detection positions are collected before the execution start time, arranged in adjacent sampling times, forming a pre-cutoff sampling window; the pre-cutoff sampling window includes at least two sampling times arranged in adjacent sampling times. The pre-cutoff sampling window includes the remaining connection region number, stress detection position number, sampling time, and corresponding real-time stress feedback data, used to record the stress state of each remaining connection region before the current executing candidate board segmentation path segment is cut off.

[0030] After the currently executing candidate board segment is cut, the execution termination time of the candidate board segment is read, and the candidate board segment is written into the executed board segment data. The corresponding connection area of ​​the candidate board segment is read according to the path connection table, and this connection area is identified as the cut connection area. The cut connection area is then removed from the set of remaining connection areas, forming the set of remaining connection areas after cutting. Subsequently, for the connection areas that still belong to the set of remaining connection areas after cutting, the written stress detection positions are read, and real-time stress feedback data of these stress detection positions are collected after the execution termination time, arranged adjacently by sampling time, forming a post-cutting sampling window. The post-cutting sampling window includes at least two sampling times arranged adjacently by sampling time. The calculation object of the cutting contribution is the connection area that still belongs to the set of remaining connection areas after the currently executing candidate board segment is cut. The post-cutting sampling window includes the remaining connection area number, stress detection position number, sampling time, and corresponding real-time stress feedback data, used to record the stress state of each remaining connection area after the currently executing candidate board segment is cut.

[0031] After forming the pre-cut sampling window and the post-cut sampling window, for the same remaining connection area, real-time stress feedback data of each stress detection position corresponding to the remaining connection area are read at the same sampling time, and these real-time stress feedback data are averaged according to the number of stress detection positions to form the regional stress feedback value at that sampling time.

[0032] After generating the regional stress feedback values ​​for each sampling time, the difference between regional stress feedback values ​​at adjacent sampling times is calculated according to the order of the sampling times, and this difference is determined as the regional stress change. The time interval between adjacent sampling times is then read, and the regional stress change is divided by the corresponding time interval to form the regional stress change rate. This rate is then written to the later sampling time in the sequence between two adjacent sampling times. The time interval is in seconds, and the regional stress change rate is in seconds (the unit of real-time stress feedback data). The regional stress change amount records the magnitude of stress feedback change in the same remaining connection area between adjacent sampling times, while the regional stress change rate records how quickly this magnitude changes relative to the sampling time interval.

[0033] Finally, the regional stress feedback value, regional stress change amount, regional stress change rate, and sampling time are written into the data fields of the corresponding remaining connection area to form a regional stress record. The regional stress record includes the regional stress record corresponding to the sampling window before cutting and the regional stress record corresponding to the sampling window after cutting. Subsequently, when forming the cutting contribution for the completed plate-splitting path segment, the difference in regional stress feedback value and the difference in regional stress change rate can be directly calculated based on the regional stress records before and after cutting the completed plate-splitting path segment, thus continuing the process of forming the cutting contribution in step three.

[0034] Step 3: For the completed plate-splitting path segment, calculate the difference in regional stress feedback value and the difference in regional stress change rate based on the regional stress records before and after cutting, and determine the path interval of the connection area according to the path connection table to form the cutting contribution. Specifically, in this embodiment, after writing the executed board-splitting path data into the completed board-splitting path segment, for each remaining connection area in the set of remaining connection areas after the cut, the regional stress records before and after the cut of the completed board-splitting path segment are read. The regional stress records before the cut are derived from the sampling window formed before the start time of the execution of the completed board-splitting path segment, and the regional stress records after the cut are derived from the sampling window formed after the end time of the execution of the completed board-splitting path segment. For the same remaining connection area, the regional stress feedback values ​​arranged adjacently according to the sampling time are first read from the regional stress records before the cut, and the difference between the regional stress feedback values ​​between adjacent sampling times is calculated; then, the absolute values ​​of the differences between the regional stress feedback values ​​at each adjacent sampling time are taken and averaged to form the reference disturbance amount before the cut. The reference disturbance amount before the cut is used to record the stress fluctuation amplitude that already exists in the remaining connection area before the actual cut of the board-splitting path segment.

[0035] After establishing the pre-cutting reference disturbance, the stress feedback value corresponding to the last sampling moment in the sampling window before the cut of the completed slab segment is read and determined as the pre-cutting regional stress feedback value. The stress feedback value corresponding to the first sampling moment in the sampling window after the cut of the completed slab segment is read and determined as the post-cutting regional stress feedback value. The difference between the post-cutting regional stress feedback value and the pre-cutting regional stress feedback value is determined as the difference between the pre- and post-cutting regional stress feedback values. The absolute value of the difference between the pre- and post-cutting regional stress feedback values ​​is subtracted from the pre-cutting reference disturbance. When the subtraction result is less than 0, the effective regional stress increment value is determined as 0; when the subtraction result is greater than or equal to 0, the subtraction result is determined as the effective regional stress increment value. Subsequently, the direction of stress increment in the effective area is determined based on the magnitude of the difference between the stress feedback values ​​before and after the cutting: when the difference between the stress feedback values ​​before and after the cutting is greater than 0, the direction of stress increment in the effective area is determined as the direction of stress increase; when the difference between the stress feedback values ​​before and after the cutting is less than 0, the direction of stress increment in the effective area is determined as the direction of stress decrease; when the difference between the stress feedback values ​​before and after the cutting is equal to 0, neither the direction of stress increase nor the direction of stress decrease is formed, and the value of stress increment in the effective area is determined as 0.

[0036] After establishing the effective stress increment values ​​and directions for the effective region, the stress change rate corresponding to the last sampling moment in the sampling window before the cut-off of the completed plate segment is read and determined as the stress change rate before cut-off. The stress change rate corresponding to the second sampling moment in the sampling window after the cut-off of the completed plate segment is read and determined as the stress change rate after cut-off. The difference between the stress change rate before and after cut-off is determined as the stress change rate difference. The direction of the stress change rate difference is determined based on the magnitude of the difference: when the difference is greater than 0, it is determined as the rate increasing direction; when it is less than 0, it is determined as the rate decreasing direction; when the difference is equal to 0, neither a rate increasing nor a rate decreasing direction is established. The direction of the difference in the rate of change of regional stress is used to record the direction of the change in the rate of stress change in the remaining connection area before and after the cut-off of the plate-splitting path segment, and is used together with the direction of the effective area stress increment to form a combined calculation result.

[0037] After determining the direction of the difference in the rate of change of regional stress, the path interval of the connection area is determined according to the path connection table. Specifically, the connection area corresponding to the completed plate-splitting path segment in the path connection table is read, and the candidate plate-splitting path segment corresponding to the current remaining connection area in the path connection table is read; the path interval of the connection area is formed based on the absolute value of the difference between the order of the corresponding path segments. The path interval of the connection area is used to record the order distance between the connection area corresponding to the completed plate-splitting path segment and the current remaining connection area on the initial plate-splitting path, and participates in the calculation of the attenuation of the subsequent cutting contribution.

[0038] Before generating the combined calculation results, the absolute value of the difference in stress change rates before and after the cutting is first scaled. Specifically, the sampling time interval used by the sampling windows before and after the cutting is read, and the absolute value of the difference in stress change rates before and after the cutting is multiplied by the sampling time interval to form the rate-equivalent stress change. The rate-equivalent stress change is the calculated value of the absolute value of the difference in stress change rates before and after the cutting after scale unification. It uses the same stress feedback unit as the effective area stress increment value and is used as the calculated value of the absolute value of the difference in stress change rates before and after the cutting after scale unification in the combined calculation results.

[0039] After determining the effective region stress increment value, effective region stress increment direction, region stress change rate difference direction, rate equivalent stress change, and connection zone path interval, a cutoff contribution is generated according to different calculation scenarios: when the effective region stress increment value is 0, the cutoff contribution is determined to be 0; when the effective region stress increment value is greater than 0 and the difference in region stress change rate before and after cutoff is 0, the effective region stress increment value is determined as the combined calculation result, and the combined calculation result is divided by the value 1 plus the connection zone path interval to form the cutoff contribution; when the effective region stress increment value is greater than 0 and the effective region stress increment direction and the region stress change rate difference direction are both... When the values ​​are consistent, the effective region stress increment value and the rate equivalent stress change value are added together to form the same-direction cutoff gain value. The same-direction cutoff gain value is determined as the combined calculation result. The combined calculation result is divided by the value of 1 plus the path interval of the connection area to form the cutoff contribution value. When the effective region stress increment value is greater than 0 and the direction of the effective region stress increment value is opposite to the direction of the difference in the region stress change rate, the absolute value of the difference between the effective region stress increment value and the rate equivalent stress change value is determined as the reverse cutoff cancellation value. The reverse cutoff cancellation value is determined as the combined calculation result. The combined calculation result is divided by the value of 1 plus the path interval of the connection area to form the cutoff contribution value.

[0040] The resulting cutoff contribution is written into the cutoff contribution record between the completed slab segment and the corresponding remaining connection area. The cutoff contribution record includes the completed slab segment number, the remaining connection area number, the baseline disturbance before cutoff, the effective area stress increment value, the effective area stress increment direction, the direction of the difference in the regional stress change rate, the rate-equivalent stress change, the connection area path interval, the combined calculation result, and the cutoff contribution. In subsequent step four, when temporarily removing the corresponding connection areas to be cut from subsequent slab segment segments and forming a set of virtual connection areas, this cutoff contribution is used to form the virtual transfer amount for each virtual connection area.

[0041] Step 4: Read the subsequent board splitting path segments that have not been written with executed board splitting path data from the candidate board splitting path segment set, temporarily remove the corresponding connection areas to be cut from the remaining connection area set to form a virtual connection area set, and form a virtual transfer amount based on the cutting contribution amount. Form the board splitting release risk value based on the virtual transfer amount and the regional stress record. Specifically, in this embodiment, after the cutoff contribution is formed, candidate board-segment path segments that have not been written with executed board-segment path data are read from the candidate board-segment path segment set and identified as subsequent board-segment path segments. The connection area to be cut off corresponding to this subsequent board-segment path segment is read according to the path connection table. Without controlling the board-segment execution mechanism to actually cut off the connection area to be cut off, a temporary removal process is performed on the current remaining connection area set, removing the connection area to be cut off from the remaining connection area set, forming a virtual connection area set corresponding to this subsequent board-segment path segment. When the virtual connection area set is empty, the board-segment release risk value corresponding to this subsequent board-segment path segment is determined to be 0, and when a risk ranking sequence is subsequently formed, this subsequent board-segment path segment and its corresponding board-segment release risk value are written into the risk ranking sequence. When the virtual connection area set is not empty, this virtual connection area set is used to record the connection areas that remain connected after the subsequent board-segment path segment is assumed to be executed, providing calculation objects for the formation of subsequent virtual transfer amounts and board-segment release risk values.

[0042] When the virtual connection region set is not empty, the boundary of the connection region to be severed before its temporary removal is read, and this boundary is determined as the original connection region boundary. Subsequently, virtual connection regions with overlapping boundary segments with the original connection region boundary are read from the virtual connection region set, and these read virtual connection regions are determined as transfer entry connection regions. When no virtual connection region with overlapping boundary segments with the original connection region boundary is read, the virtual connection region whose path segment sequence is adjacent to the path segment sequence of the connection region to be severed is read according to the path connection table, and this read virtual connection region is determined as the transfer entry connection region. The transfer entry connection region is used to record the virtual connection region that first receives the impact of the original connection region boundary release after the connection region to be severed is temporarily removed.

[0043] If there are more than two transfer entry connection areas, the length of the overlapping boundary segment between each transfer entry connection area and the original connection area is retained and included in the calculation when forming the virtual boundary openness. After forming the transfer entry connection area, for each virtual connection area, each transfer entry connection area is used as the first path level. Based on the adjacency relationship of the path segments between connection areas in the path connection table, the system expands outward to obtain the candidate path levels of the virtual connection area relative to each transfer entry connection area. The candidate path level with the smallest value is determined as the path level of the virtual connection area. When there is only one transfer entry connection area, it is directly used as the first path level. Based on the adjacency relationship of the path segments between connection areas in the path connection table, the system expands outward to determine the path level of each virtual connection area. The path level is used to record the connection position relationship of each virtual connection area relative to the transfer entry connection area.

[0044] After determining the path hierarchy, for each virtual connection region in the set of virtual connection regions, the boundary adjacency length between the virtual connection region and each severed connection region is read, and the read boundary adjacency lengths are accumulated to form the severed boundary adjacency length; simultaneously, the boundary adjacency length between the virtual connection region and the original connection region boundary is read. Then, the severed boundary adjacency length is added to the boundary adjacency length between the virtual connection region and the original connection region boundary, and the ratio between the sum and the connection region boundary length of the virtual connection region is determined as the virtual boundary openness. The virtual boundary openness is used to record the degree of exposure of the virtual connection region relative to the open boundary and the original connection region boundary after the temporary removal of the connection region to be severed.

[0045] The preset hierarchical correction relationship includes the correspondence between path levels and hierarchical correction values. Specifically, the preset hierarchical correction relationship records the hierarchical correction values ​​corresponding to the first path level to the maximum allowed path level respectively; wherein, the hierarchical correction value corresponding to the first path level is used to record the direct acceptance ratio of the transfer entry connection area to the cutoff contribution, and the hierarchical correction values ​​corresponding to the second path level to the maximum allowed path level are used to record the acceptance ratio after being transferred outward along the path segment sequential adjacency relationship. When the path level corresponding to the virtual connection area exceeds the maximum allowed path level, the hierarchical correction value corresponding to the virtual connection area is determined to be 0. According to the path level corresponding to each virtual connection area, the corresponding hierarchical correction value is read from the preset hierarchical correction relationship. When multiple completed board-segmenting path segments are read to the cutoff contribution of the same virtual connection area, each cutoff contribution is read according to the number of the completed board-segmenting path segment, and the cutoff contributions are accumulated to form the cutoff contribution to be transferred; when only one completed board-segmenting path segment is read to the cutoff contribution of the virtual connection area, the cutoff contribution is determined as the cutoff contribution to be transferred. The virtual transfer amount of the virtual connection zone is formed by multiplying the contribution amount to be cut off, the virtual boundary openness corresponding to the virtual connection zone, and the hierarchical correction value.

[0046] After generating the virtual transfer amount for each virtual connection zone, for each virtual connection zone in the set of virtual connection zones, the regional stress feedback value corresponding to the sampling time adjacent to the temporary removal time of the subsequent board segment is read from the regional stress record, and this regional stress feedback value is determined as the current regional stress feedback value. The temporary removal time of the subsequent board segment is the risk assessment time corresponding to the calculation of the temporary removal of the connection zone to be cut for that subsequent board segment, not the actual cutting time of the board execution mechanism. Subsequently, the current regional stress feedback value is added to the virtual transfer amount corresponding to the virtual connection zone to form the virtual post-execution regional stress prediction value for that virtual connection zone. The virtual post-execution regional stress prediction value is used to record the predicted stress state of the virtual connection zone after the assumed execution of the subsequent board segment, based on the current regional stress feedback and the effects of the transfer.

[0047] After generating the predicted stress values ​​for each virtual connection region, the maximum and average predicted values ​​are read. A prediction set difference is generated based on the difference between the maximum and average predicted values. This prediction set difference records the concentration of predicted stress values ​​within the set of virtual connection regions. Next, the virtual transfer amounts corresponding to each virtual connection region are read. When the sum of all virtual transfer amounts is 0, the transfer concentration ratio is set to 0. When the sum of all virtual transfer amounts is greater than 0, a transfer concentration ratio is generated based on the ratio of the maximum virtual transfer amount to the sum of all virtual transfer amounts. This transfer concentration ratio records the concentration of virtual transfer amounts within the set of virtual connection regions.

[0048] After generating the predicted concentration difference and the transfer concentration ratio, the predicted concentration difference is multiplied by the transfer concentration ratio to form the plate release risk value corresponding to the subsequent plate-slab path segment. The plate release risk value is used to record the release risk formed by the combined regional stress prediction concentration and virtual transfer amount concentration in the virtual connection area set after the subsequent plate-slab path segment is assumed to be executed. The subsequent step five reads the plate release risk value corresponding to each subsequent plate-slab path segment to determine the next execution path segment.

[0049] Step 5: Determine the next execution path segment based on the risk value of the board release corresponding to each subsequent board segment, form the optimized board segment path, and control the board execution mechanism to execute the subsequent board segments according to the optimized board segment path; Specifically, in this embodiment, after the risk value for each subsequent board release path segment is formed, the subsequent board release path segments in the candidate board release path segment set that have not yet been written with executed board release path data are read, and the corresponding board release risk value for each subsequent board release path segment is read. The subsequent board release path segments are sorted from low to high according to their board release risk values ​​to form a risk ranking sequence. The risk ranking sequence includes the subsequent board release path segment number, the corresponding board release risk value, and the ranking position. The process of forming a risk ranking sequence and determining the next execution path segment is called the path selection process; each time the path selection process is executed, the risk ranking sequence corresponding to this path selection process and the ranking position of each subsequent board release path segment are written into the path selection record.

[0050] After the path selection record is generated, the sorting position of the same subsequent sub-segment path segment in a preset number of consecutive path selection processes is read. When the same subsequent sub-segment path segment is within a preset number of sorting positions at the end of the risk sorting sequence in a preset number of consecutive path selection processes, the subsequent sub-segment path segment is determined as a risk retention path segment. The preset number of sorting positions at the end is used to limit the number of sorting positions near the end in the risk sorting sequence; for example, when the risk sorting sequence contains ten subsequent sub-segment path segments, if the preset number of sorting positions at the end is set to three, then the subsequent sub-segment path segments at the eighth, ninth, and tenth positions are within the preset number of sorting positions at the end. Therefore, the risk retention path segment is not formed by a single risk sorting result, but is determined by the sorting positions in multiple path selection processes.

[0051] After a risk-stuck path segment is formed, the corresponding connection area to be severed is read. Then, from the remaining subsequent board-level path segments, subsequent board-level path segments that have at least one of the following relationships with the connection area to be severed: boundary adjacency and path segment order adjacency, are read and written into the set of retreat-associated path segments. Boundary adjacency is determined based on whether there is an adjacent boundary position between the connection area to be severed corresponding to the retreat-associated path segment and the connection area to be severed corresponding to the risk-stuck path segment. Path segment order adjacency is determined based on whether the path segment order of the corresponding candidate board-level path segments in the path connection table is adjacent. The set of retreat-associated path segments records subsequent board-level path segments that can have a retreat effect on the adjacent connection state of the risk-stuck path segment through prior execution.

[0052] For each retreat-related path segment in the set of retreat-related path segments, without controlling the actual severing of the corresponding connection area to be severed by the slab execution mechanism, the corresponding connection area to be severed is temporarily removed from the set of remaining connection areas. Then, according to the aforementioned slab release risk value formation method, the slab release risk value corresponding to the risk-delayed path segment is re-formed. This re-formation process is only used for retreat evaluation during the path selection process and does not correspond to actual slab execution. Subsequently, the original slab release risk value of the risk-delayed path segment is read, and the difference between the original slab release risk value and the re-formed slab release risk value is calculated. When the difference is less than 0, the risk retreat difference is determined to be 0; when the difference is greater than or equal to 0, the difference is determined to be the risk retreat difference. The risk retreat difference is used to record the degree of reduction in the slab release risk value of the risk-delayed path segment after the retreat-related path segment is assumed to be executed first.

[0053] After generating the risk concession difference, the starting positions of the concession-related path segment and the risk-delayed path segment are read, and the path length between the two starting positions is read along the initial board separation path to form the initial path switching length. Then, a preset path reference length is read, and the result of dividing the initial path switching length by the preset path reference length is determined as the path switching distance. The preset path reference length includes the average path length of each candidate board separation path segment in the candidate board separation path segment set; in another embodiment, the preset path reference length includes the average path length between the median positions of adjacent connection area boundaries in the PCB panel structure data. Therefore, the path switching distance is a dimensionless value used to record the path switching cost between the concession-related path segment and the risk-delayed path segment.

[0054] After determining the path switching distance, a retreat release amount is generated based on the risk retreat difference and the path switching distance. Specifically, the retreat release amount is formed by dividing the risk retreat difference by 1 and adding the path switching distance. The risk retreat difference is used to record the degree of risk reduction in the risk-delayed path segment, and the path switching distance is used to record the path switching cost. Therefore, the retreat release amount records both the risk reduction magnitude and the path switching cost.

[0055] After determining the release amount corresponding to each yield-related path segment, the yield-related path segments in the set are sorted from highest to lowest according to the release amount, and the yield-related path segment with the highest release amount is determined as the next execution path segment. When no risk retention path segment is formed, the subsequent board release risk value ranking first is directly determined as the next execution path segment based on the risk ranking sequence; when a risk retention path segment is formed and the set of yield-related path segments is empty, the subsequent board release risk value ranking first is directly determined as the next execution path segment based on the risk ranking sequence. After determining the next execution path segment, the optimized board release path is updated according to the next execution path segment, and the board release execution mechanism is controlled to execute the subsequent board release according to the optimized board release path; after the next execution path segment is cut off, it is written into the executed board release path data, and the next path selection process begins.

[0056] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A dynamic optimization method for PCB depaneling paths based on real-time stress feedback, characterized in that, include: Step 1: Obtain PCB panel structure data, initial board separation path data, and stress detection location data. Read the connection area location data from the PCB panel structure data and form a connection area set. Divide the candidate board separation path segment set according to the initial board separation path data, and establish a path connection table based on the positional correspondence between the cutting coverage of the candidate board separation path segments and the connection area set. Step 2: During the plate separation process, obtain real-time stress feedback data and the data of the plate separation path that has been executed. Remove the cut connection areas from the connection area set according to the path connection table to form the remaining connection area set. Then, based on the stress detection location data and real-time stress feedback data, form the regional stress record of each remaining connection area. Step 3: For the completed plate-splitting path segment, calculate the difference in regional stress feedback value and the difference in regional stress change rate based on the regional stress records before and after cutting, and determine the path interval of the connection area according to the path connection table to form the cutting contribution. Step 4: Read the subsequent board splitting path segments that have not been written with executed board splitting path data from the candidate board splitting path segment set, temporarily remove the corresponding connection areas to be cut from the remaining connection area set to form a virtual connection area set, and form a virtual transfer amount based on the cutting contribution amount. Form the board splitting release risk value based on the virtual transfer amount and the regional stress record. Step 5: Determine the next execution path segment based on the risk value of each subsequent board segment release, forming an optimized board segment path, and control the board segment execution mechanism to execute the subsequent board segments according to the optimized board segment path.

2. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 1, characterized in that, In step one, after forming the connection zone set and establishing the path connection table, stress detection locations are written to the connection zones in the connection zone set, including: The boundary of each connection zone is determined based on the location data of the connection zone, and the midpoint of the connection zone boundary is formed based on the boundary of the connection zone. Based on the path connection table, read the candidate board path segment corresponding to each connection area, and form the cutting extension baseline by the cutting coverage of the candidate board path segment; Read the clamping constraint area from the PCB panel structure data, and form a clamping force transmission baseline based on the midpoint of the connection area boundary and the boundary of the clamping constraint area closest to the connection area; With the midpoint of the boundary of the connecting area as the origin, and the direction of the cutting extension baseline and the direction of the clamping force transmission baseline as the first coordinate direction and the second coordinate direction respectively, a local belonging coordinate system is formed. The stress detection location is mapped to the local assigned coordinate system to form the cut-off direction projection coordinates and the clamping direction projection coordinates. Adjacent connection zones are determined based on their positional adjacency relationship, and ownership boundaries are formed based on the corresponding cutting direction projection coordinates and clamping direction projection coordinates of adjacent connection zones. The stress detection domain is enclosed by the boundary line. The stress detection locations that fall within the stress detection domain are written into the data fields of the corresponding connection area, forming a set of connection areas with stress detection locations written in them.

3. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 1, characterized in that, In step one, the candidate segmentation path set is divided based on the initial segmentation path data, including: The boundary of the connection area is determined based on the location data of the connection area, and the path extension direction and cut-off coverage area are determined based on the initial board path data; Along the path extension direction, the first position where the cut-off coverage enters the boundary of the connecting area is determined as the entry point to the side boundary, and the last position where the cut-off coverage leaves the boundary of the connecting area is determined as the exit point from the side boundary. The cut-off coverage direction and cut-off coverage length are determined by the entry and exit points of the side boundary, and written into the corresponding connection area to form cut-off coverage morphology data; Arrange the cut-over pattern data according to the order of the initial board path covering the connection area to form a cut-over pattern sequence; The direction of the cut-off coverage length difference is formed based on the cut-off coverage length difference between adjacent connected regions. Based on the entry and exit side boundary positions of each connection area, the cut-off coverage area corresponding to each connection area is divided into a candidate board path segment, so that each candidate board path segment corresponds to a connection area. Based on whether the path interval between adjacent connecting areas covers other connecting areas, whether the projection direction of the cut-off coverage direction of adjacent connecting areas on the path extension direction is consistent, and the direction of the cut-off coverage length difference between adjacent connecting areas, the path segment order adjacency relationship between adjacent candidate board path segments is determined. A set of candidate board path segments is formed according to the order in which the initial board path covers the connecting area and the sequential adjacency relationship between adjacent candidate board path segments.

4. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 2, characterized in that, In step two, regional stress records for each remaining connection zone are generated based on stress detection location data and real-time stress feedback data, including: Before executing the current candidate board segment, read the execution start time of the candidate board segment, and collect the real-time stress feedback data of the stress detection positions written in each remaining connection area before the execution start time, arranged in adjacent order of sampling time, to form a pre-cut-off sampling window; After the candidate board segment being cut off is completed, the execution termination time of the candidate board segment is read, and the real-time stress feedback data of the stress detection positions written in each remaining connection area are collected after the execution termination time and arranged in adjacent order of sampling time to form a sampling window after cutting off. For the same remaining connection area, the real-time stress feedback data corresponding to each stress detection position at the same sampling time are averaged according to the number of stress detection positions to form the regional stress feedback value at that sampling time. The regional stress change is formed based on the difference in regional stress feedback values ​​between adjacent sampling times, and the regional stress change rate is formed based on the time interval between the regional stress change and adjacent sampling times. The regional stress feedback value, regional stress change amount, regional stress change rate, and sampling time are written into the corresponding remaining connection area to form a regional stress record.

5. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 4, characterized in that, The methods for determining the cutoff contribution in step three include: Read the difference in regional stress feedback values ​​between adjacent sampling times from the regional stress record before the cut-off of the completed plate-segment path segment, and average the absolute values ​​of the regional stress feedback value differences to form the reference disturbance amount before the cut-off. Subtract the reference disturbance before cutting from the absolute value of the difference between the stress feedback values ​​of the region before and after cutting. If the subtraction result is less than 0, the effective stress increment value is determined to be 0. If the subtraction result is greater than or equal to 0, the subtraction result is determined to be the effective stress increment value. The direction of the effective stress increment is determined according to the direction of change of the difference between the stress feedback values ​​of the region before and after cutting. The direction of the difference in the rate of change of stress in the region is determined based on the direction of the numerical change of the difference in the rate of change of stress in the region before and after the cut. When the effective stress increment is 0, the cutoff contribution is determined to be 0. When the effective area stress increment value is greater than 0, the combined calculation result of the effective area stress increment value and the absolute value of the difference in the area stress change rate before and after the cut is determined based on whether the difference in the area stress change rate before and after the cut is 0 and whether the direction of the effective area stress increment is consistent with the direction of the difference in the area stress change rate. The combined calculation result is then divided by the value 1 plus the path interval of the connection area to form the cut contribution.

6. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 1, characterized in that, Step four, the method for forming a virtual transfer amount based on the cutoff contribution, includes: After temporarily removing the connection area to be cut from the remaining connection area set and forming a virtual connection area set, the connection area boundary of the connection area to be cut before temporary removal is read as the original connection area boundary. Read virtual connection regions from the set of virtual connection regions that have boundary segments that overlap with the original connection region boundary, and form the transfer entry connection region; Taking the transfer entry connection area as the first path level, the candidate sub-board path segments corresponding to each virtual connection area are read according to the path connection table, and the path level of each virtual connection area in the virtual connection area set is determined according to the path segment order adjacency relationship between the candidate sub-board path segments. For each virtual connection region, the boundary adjacency length between the virtual connection region and each cut connection region is read and accumulated to form the cut boundary adjacency length. The boundary adjacency length between the virtual connection region and the original connection region boundary is read. The virtual boundary openness is formed by the ratio between the sum of the cut boundary adjacency length and the boundary adjacency length between the virtual connection region and the original connection region boundary and the connection region boundary length of the virtual connection region. Based on the path level, the corresponding level correction value is read from the preset level correction relationship that records the correspondence between path level and level correction value, and the virtual transfer amount of the virtual connection area is formed based on the cutoff contribution, virtual boundary openness and level correction value.

7. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 1, characterized in that, Step four, the method for generating the plate release risk value based on the virtual transfer amount and regional stress record, includes: For each virtual connection region in the set of virtual connection regions, the regional stress feedback value corresponding to the sampling time adjacent to the temporary removal time of the subsequent plate path segment is read from the regional stress record and used as the current regional stress feedback value. The current regional stress feedback value is added to the virtual transfer amount corresponding to the virtual connection region to form the predicted regional stress value after virtual execution. Read the maximum and average predicted values ​​from the predicted values ​​of the virtual post-execution region corresponding to each virtual connection area, and form a prediction set difference based on the difference between the maximum and average predicted values; Read the virtual transfer amount corresponding to each virtual connection area. When the sum of all virtual transfer amounts is 0, the transfer concentration ratio is set to 0. When the sum of all virtual transfer amounts is greater than 0, the transfer concentration ratio is formed based on the ratio of the maximum virtual transfer amount to the sum of all virtual transfer amounts. Multiply the predicted concentration difference by the transfer concentration ratio to form the risk value of the board release.

8. The PCB depaneling path dynamic optimization method based on real-time stress feedback according to claim 7, characterized in that, In step five, the next execution path segment is determined based on the risk value of the board release corresponding to each subsequent board segment, including: A risk sorting sequence is formed based on the risk value of each subsequent board segment release, and the process of forming a risk sorting sequence and determining the next execution path segment is called the path selection process. The same subsequent board path segment that is within the preset number of consecutive preset times at the end of the risk sorting sequence is identified as the risk retention path segment. Subsequent board-segmented path segments that have at least one of the following relationships with the risk-delayed path segment corresponding to the area to be cut off: boundary adjacency relationship and path segment sequence adjacency relationship; are written into the set of retreat-associated path segments. For each yielding associated path segment, after temporarily removing the corresponding connection area to be cut from the set of remaining connection areas, the risk release risk value of the risk retention path segment is re-formed according to claim 7, and a risk yielding difference is formed based on the original risk release risk value of the risk retention path segment and the re-formed risk release risk value. The path switching distance is formed based on the path length along the initial board path between the starting positions of the retreat-related path segment and the risk retention path segment, and the retreat release amount is formed based on the risk retreat difference and the path switching distance. The yield-related path segment that ranks first in yield release amount is determined as the next execution path segment.

Images