A monitoring and control system for the assembly and folding process of a rigid-flex printed circuit board

Abstract

The application discloses a soft and hard combined plate assembly winding and folding process monitoring control system, relates to the soft and hard combined plate assembly winding and folding process monitoring field, and comprises the following steps: strain sensor groups are arranged on the surface of the soft and hard combined plate, stress sensors are densely arranged in the soft and hard joint area, the plate body coordinate mapping relationship associated with the motion coordinate system is established, and strain rate and stress distribution data are collected in real time; based on the historical data sequence analysis abnormal change trend of continuous multiple collection periods, the dynamic risk assessment is carried out in combination with the stress change gradient and the moving trend of the stress concentration area; for the potential fracture point and the risk level positioned, different priority control areas are divided according to the spatial distribution density, and the winding and folding execution mechanism is controlled to execute the differentiated parameter adjustment strategy, so that the soft plate fracture in the production process and the assembly winding of the soft and hard combined plate is avoided.

Description

Technical Field

[0001] This invention belongs to the field of monitoring technology for the folding process of rigid-flex PCB assembly, and relates to a monitoring and control system for the folding process of rigid-flex PCB assembly. Background Technology

[0002] Rigid-flex boards, as key circuit board components combining the support of rigid boards with the flexibility of flexible boards, are widely used in modern three-dimensional assembled electronic devices. Bending is a crucial step in their assembly, where the flexible portion is bent into a predetermined shape to achieve three-dimensional assembly. However, due to the abrupt change in mechanical properties at the rigid-flex interface of the material, and the high risk of stress concentration during bending, microcracks can easily develop, leading to the breakage of flexible circuits and severely impacting product yield.

[0003] Currently, although the technology for monitoring the bending process of rigid-flex PCBs has been applied to some extent, it still has the following shortcomings in practical applications: First, the existing technology relies heavily on preset process parameters and offline sampling inspection, and uses fixed threshold comparison for anomaly judgment. It cannot capture the instantaneous stress anomalies that occur in the rigid-flex PCB area during the bending process, so the mechanical signal changes in the early stage of microcrack initiation are completely ignored, resulting in the inability to detect the risk of fracture in time.

[0004] Secondly, the risk assessment logic of existing technologies is usually based on static mechanical assumptions to build risk assessment models. However, the bending process itself is a dynamic process, and the material behavior exhibits obvious nonlinear characteristics. Static models are difficult to accurately describe this process. At the same time, risk assessment often relies on data at isolated time points or fixed safety thresholds, and cannot track the continuous change trend of data during the production of rigid-flex PVC boards. This leads to delayed risk prediction, inaccurate problem location, and a high risk of misjudgment or underreporting.

[0005] Finally, existing technologies only passively alarm or perform uniform parameter adjustments for abnormal results detected. Once a high-risk fracture point occurs during the production of rigid-flex PCBs, it is impossible to actively intervene to disperse stress and suppress crack propagation in real time, leading to the propagation of microcracks or even the fracture of the flexible board, which affects the final product qualification rate. Summary of the Invention

[0006] In view of this, in order to solve the problems mentioned in the background art, the present invention provides a monitoring and control system for the bending process of rigid-flex PCB assembly.

[0007] The objective of this invention can be achieved through the following technical solution: a monitoring and control system for the assembly and bending process of a rigid-flex PCB, comprising: a data acquisition module for real-time acquisition of strain rate and stress distribution data of the rigid-flex PCB during the bending process.

[0008] The anomaly comparison module is used to compare strain rate and stress distribution data with standard process data to determine the abnormal change rate.

[0009] The stress concentration location module is used to determine the density of abnormal points based on the abnormal change rate, locate the stress concentration area based on the density of abnormal points, and determine the stress change gradient and movement trend of the stress concentration area.

[0010] The fracture risk assessment module is used to calculate the stress peak at the soft-hard interface based on the stress change gradient and movement trend, locate potential fracture points based on the stress peak, and assess the risk level.

[0011] The stress control module is used to control the bending actuator to adjust the bending radius and speed in real time based on potential fracture points and risk levels, and to construct a stress distribution map.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention arranges strain sensor groups on the surface of the soft and hard bond plate and densifies stress sensors in the soft and hard interface area, establishes a plate coordinate mapping relationship with the motion coordinate system, and collects strain rate and stress distribution data in real time. This solves the problem of relying on preset parameters and offline sampling and being unable to capture instantaneous stress anomalies, and realizes real-time perception of the mechanical state of the entire bending process, laying a reliable data foundation for early identification of microcrack initiation and fracture risk warning.

[0013] (2) Based on the analysis of the abnormal change trend of historical data sequence from multiple consecutive acquisition cycles, this invention combines the stress change gradient and movement trend of stress concentration area to carry out dynamic risk assessment. This solves the problem that dynamic nonlinear behavior cannot be described due to reliance on static mechanical assumptions and isolated time point data, making it difficult to make early predictions. It realizes dynamic assessment and accurate positioning of potential fracture risks, significantly improving the accuracy of risk judgment.

[0014] (3) The present invention divides the potential fracture points and their risk levels into different priority control areas according to their spatial distribution density, and controls the bending actuator to perform differentiated parameter adjustment strategies, thereby realizing real-time stress control of high-risk areas during the bending process. This can effectively disperse stress concentration and suppress microcrack propagation, thereby directly avoiding soft plate fracture and ensuring the qualification rate of the final product. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram showing the connections of the various modules in the system of the present invention.

[0017] Figure 2 This is a flowchart of the stress concentration region identification and boundary determination process of the present invention.

[0018] Figure 3 This is a flowchart of the priority control of potential fracture points and the adjustment of bending parameters in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 As shown, this invention provides a monitoring and control system for the bending process of rigid-flex PCB assembly, including a data acquisition module, an anomaly comparison module, a stress concentration location module, a fracture risk assessment module, and a stress control module. All modules are connected in the order described above.

[0021] The data acquisition module is used to collect data on strain rate and stress distribution of the rigid-flexible plate during the bending process in real time.

[0022] Due to the significant differences in mechanical properties between the rigid and flexible materials in rigid-flex boards, especially in the interface area, the stress-strain distribution is extremely complex during bending. Any minute, localized excessive deformation can eventually lead to the initiation and propagation of microcracks, causing irreversible damage to the product. Because these localized abnormal changes cannot be detected in real time, process control becomes lagging, leaving only the passive acceptance of defective products.

[0023] Based on this, the above-mentioned strain rate and stress distribution data acquisition steps are as follows: First, strain sensor groups are arranged on the surface of the soft-hard bonded plate at a preset base spacing, and stress sensors are arranged symmetrically on both sides of the boundary line in the soft-hard interface area at a spacing smaller than the base spacing.

[0024] It should be added that a dynamic foil strain sensor can be selected for the strain sensor, which can adapt to the dynamic deformation of the flexible and rigid bonding plate during bending, and avoid data distortion caused by exceeding the range; a miniature piezoresistive stress sensor can be selected for the stress sensor, to avoid mutual interference between adjacent sensors due to excessive sensor size.

[0025] The preset spacing between foundations is determined based on the plate area and shape. Larger plate areas and more complex shapes require appropriately reduced spacing to avoid blind spots in data acquisition. Smaller plate areas and more regular shapes allow for a more balanced approach, increasing the spacing to optimize data accuracy and acquisition costs. The same principle applies to the spacing of stress sensors. This achieves full-area data coverage of the plate and enhances data acquisition accuracy in high-risk areas, laying the foundation for subsequent real-time acquisition of accurate raw strain and stress data.

[0026] Next, establish the coordinate mapping relationship between all sensors and the motion coordinate system of the bending actuator.

[0027] Specifically, three non-collinear coordinate reference points are marked at the corners of the plate. The coordinates of the reference points in the motion coordinate system of the actuator are measured by a laser positioning instrument. With the reference points as the origin, the X-axis is along the main conveying direction of the actuator, which corresponds to the X-axis of the plate; the Y-axis is along the lateral adjustment direction of the actuator, which corresponds to the Y-axis of the plate; and the Z-axis is along the vertical lifting direction of the actuator, which corresponds to the Z-axis of the plate. A local coordinate system of the plate is established, and the coordinates of each sensor in the local coordinate system are calculated. Then, through the global coordinate transformation of the reference points, the global coordinates of all sensors in the motion coordinate system of the actuator are obtained.

[0028] The bending actuator is activated to perform bending operations, and the deformation of each strain sensor is read in real time at continuous time points. The local strain rate is calculated based on the deformation changes at adjacent time points.

[0029] The stress values ​​measured by each stress sensor are read synchronously, and the stress values ​​at each location are fitted into stress distribution data by spatial interpolation based on their position coordinates.

[0030] It should be noted that since the stress values ​​read are discretely distributed, spatial interpolation can use the inverse distance weighted interpolation method to fit the stress values ​​at each location into continuous stress distribution data covering the entire plate surface.

[0031] The anomaly comparison module is used to compare strain rate and stress distribution data with standard process data to determine the abnormal change rate.

[0032] Because it is impossible to systematically compare the real-time acquired strain rate and stress distribution data with standard process data, it is impossible to quantify the deviation of the real-time data of each sensor point from the benchmark value, to distinguish between occasional fluctuations and continuous anomalies in the data, and to determine the rate of abnormal change that can support subsequent analysis, resulting in the inability to identify process anomalies in a timely manner.

[0033] Therefore, the specific contents of the above-mentioned anomaly comparison module are as follows: calculate the deviation of the real-time values ​​of each sensor point in the real-time strain rate data and real-time stress distribution data from the corresponding benchmark values ​​in the standard process data.

[0034] It should be added that the corresponding baseline values ​​in the standard process data are obtained through statistical analysis of historical qualified bending process data. The deviation range is determined by calculating the percentage of the absolute value of the difference between the real-time value and the corresponding baseline value to the baseline value at the corresponding point.

[0035] A real-time deviation distribution map is generated based on the deviation of all points, and data points located in the high-value area of ​​the distribution map are identified. The sensor locations corresponding to these data points are determined as initial anomaly points.

[0036] Because the greater the deviation, the greater the deviation between the real-time process state and the safety benchmark, the more likely the stress value and strain state at the corresponding point are to exceed normal fluctuations, which poses a potential risk of stress concentration or even fracture. Ultimately, the data points located in the high-value area on the distribution map are identified as initial anomaly points.

[0037] The process for determining the high-value region criteria is as follows: First, the deviation amplitude data of each sensor point under normal bending conditions of the rigid-flex PCB is extracted. After sorting these data from largest to smallest, the top 10% of the deviation amplitude values ​​in the sorted results are taken. The specific percentage can also be adjusted according to the process accuracy requirements as the high-value threshold. Then, the real-time deviation amplitude distribution map is aligned with the sensor layout. Subsequently, high-value points with real-time deviation amplitudes greater than or equal to the high-value threshold are selected. Finally, the region containing no less than three consecutive adjacent high-value points is determined as the high-value region.

[0038] The historical deviation data of each initial anomaly point within multiple consecutive acquisition cycles are traced to form a data sequence.

[0039] Finally, if the data sequence exhibits a monotonically increasing characteristic over three or more consecutive collection periods, it is identified as a valid outlier, and the average change per unit time of each valid outlier during the trend period is calculated as the outlier rate of change for each point.

[0040] It should be noted that an increase in the deviation amplitude in one or two acquisition cycles may be due to accidental factors such as temporary interference from the sensor or instantaneous minor fluctuations in the actuator; however, a monotonically increasing deviation amplitude for three or more consecutive cycles indicates that the increase in the deviation amplitude is a stable and continuous trend, excluding accidental interference, and reflects that the deviation between the real-time strain rate and stress value and the standard reference is continuously expanding, which is a real and existing abnormal development trend.

[0041] The stress concentration location module is used to determine the density of abnormal points based on the abnormal change rate, locate the stress concentration area based on the density of abnormal points, and determine the stress change gradient and movement trend of the stress concentration area.

[0042] Since the density of anomaly points is a key indicator for quantifying the degree of anomaly concentration in various regions of the rigid-flex PCB and providing a core basis for the subsequent accurate location of stress concentration areas, the specific steps for determining the density of anomaly points based on the rate of change of anomalies are as follows: First, the rigid-flex PCB is divided into multiple analysis units, the number of valid anomalies falling into each analysis unit is counted, and the arithmetic mean of the rate of change of anomalies of all valid anomalies in each analysis unit is calculated.

[0043] It should be added that the method for dividing the analysis cells is mainly based on the geometric layout of the rigid-soft interface and the sensor distribution characteristics. For example, the interface can be divided into uniform grid cells according to the location of the rigid-soft interface and the sensor density. Specifically, assuming the rigid-soft interface is a rectangular area, it can be divided into multiple square or rectangular analysis cells of the same size. Each cell covers a certain area of ​​the interface, and the cell boundaries are aligned with the sensor positions to accurately count the number of valid anomalies falling within the cell and calculate the average anomaly rate of change.

[0044] Finally, based on the product of the number of valid outliers and the average rate of change of outliers, an outlier cluster density index for each analysis unit is generated and recorded as the outlier cluster density of each analysis unit.

[0045] Considering that stress concentration is a phenomenon caused by the difference in mechanical properties of materials when flexible and rigid plates are bent, the region usually exhibits the spatial aggregation characteristics of abnormal points. The aggregation density of abnormal points can directly reflect the potential degree of stress concentration. However, isolated high-density units or instantaneous fluctuations can easily cause positioning errors, and the regional boundaries and stability need to be clearly defined to support subsequent risk assessment and control.

[0046] See Figure 2 As shown, the specific steps for locating stress concentration areas are as follows: the analysis units with anomaly point aggregation density exceeding the preset density threshold are identified as candidate units, and adjacent candidate units are determined to be adjacent to each other when they share a boundary or corner point.

[0047] The specific steps for obtaining the preset density threshold are as follows: First, during the initial debugging phase of the system, collect a large amount of abnormal point cluster density data from known qualified process cycles. Set an initial threshold by calculating the mean and standard deviation, for example, the mean plus twice the standard deviation. Then, during actual production line operation, apply the initial threshold for real-time monitoring. Whenever an early warning occurs, record the result as a false alarm or a real risk. Based on the accuracy statistics of the early warning over a period of time, if there are too many false alarms, appropriately increase the threshold to reduce sensitivity; if there are missed alarms, decrease the threshold to improve detection sensitivity. This iterative feedback and adjustment process yields the optimal preset density threshold.

[0048] Adjacent candidate elements are merged to form a continuous initial stress concentration region.

[0049] When the density of anomalous points in each preliminary stress concentration area does not decrease within three or more consecutive acquisition cycles, the area is confirmed as an effective stress concentration area.

[0050] It should be noted that the fact that the density of abnormal points does not decrease within three or more consecutive collection cycles indicates that the stress anomaly at this location has a continuous and cumulative trend, ruling out occasional interference and thus confirming that it is a stable and potentially developing real risk source.

[0051] Identify all peripheral analysis elements that constitute the effective stress concentration region, extract the coordinates of the outer boundary vertices of each peripheral analysis element, connect the outer boundary vertices in sequence to form a closed boundary polygon, and record the region enclosed by the boundary polygon as the boundary range of the stress concentration region.

[0052] Specifically, extracting the coordinates of the outer boundary vertices of each peripheral analysis unit can reflect the actual spatial contour of the stress concentration area and avoid the range definition deviation caused by the incomplete boundary of a single analysis unit. Connecting the outer boundary vertices in sequence to form a closed boundary polygon conforms to the physical characteristics of continuous stress concentration distribution, ensuring that the boundary has no gaps or overlaps, and providing a clear geometric basis for the subsequent output of the boundary coordinates and spatial location information of the effective stress concentration area.

[0053] Output the boundary coordinates and spatial location information of all effective stress concentration regions.

[0054] The steps for determining the stress change gradient and movement trend are as follows: First, determine the stress maximum and minimum points within the stress concentration area, calculate the difference vector between the coordinates of the two points, record the direction of the difference vector as the stress change gradient direction, and calculate the ratio of the stress difference between the two points to their distance to obtain the stress change gradient value.

[0055] Next, the coordinates of the stress maximum points in the current cycle and the previous two cycles are extracted. The direction of the vector pointing from the coordinates of the first cycle to the coordinates of the current cycle is taken as the direction of movement. The coordinate distance between the current cycle and the first cycle is obtained, and the ratio is calculated with the corresponding time interval to obtain the movement rate.

[0056] Finally, the direction of movement and the rate of movement are used together to determine the stress movement trend in the stress concentration area.

[0057] The fracture risk assessment module is used to calculate the stress peak at the soft-hard interface based on the stress change gradient and movement trend, locate potential fracture points based on the stress peak, and assess the risk level.

[0058] Among them, the stress peak at the soft-hard interface is the core basic data for the fracture risk assessment module to locate potential fracture points and classify risk levels. Its calculation relies on the stress change gradient and movement trend of the stress concentration area to ensure accuracy. The specific calculation process is as follows: First, draw an extension line from the stress maximum point to the soft-hard interface along the stress change gradient direction, and determine the intersection of the extension line and the soft-hard interface.

[0059] Then, when the intersection point is located within the soft-hard boundary zone, and the direction of movement of the stress concentration area points towards the soft-hard boundary zone, it is determined that the stress concentration area constitutes a trend influence.

[0060] Furthermore, the weighted sum of the maximum stress value and the stress change gradient value in the stress concentration area that constitutes the trend influence is taken as the theoretical stress peak at the soft-hard interface, with the moving rate as the weighting factor.

[0061] The formula for calculating the theoretical peak stress is as follows: .

[0062] Where P represents the theoretical peak stress at the soft-hard interface. denoted by , where represents the current maximum stress value within the stress concentration region, k represents the adjustment coefficient factor, v represents the movement rate of the stress concentration region, and G represents the stress change gradient value.

[0063] The steps for obtaining the adjustment coefficient factor are as follows: First, a large database of historical qualified and unqualified bending processes is collected. This database includes the maximum stress value, stress change gradient value, movement rate of stress concentration area, and the actual situation of whether microcracks or fractures occur at the soft-hard interface through offline detection. Then, based on the above bending process database, the actual maximum safe stress observed in the process samples that have not been damaged is used as the target value. Substitute the current maximum stress value, the moving rate of the stress concentration region, and the stress change gradient value into the formula. A reverse calculation is performed to obtain a large number of candidate k values; finally, the mode of all candidate k values ​​is taken as the final k value. After statistical analysis of the rigid-flex PCB material and conventional bending process parameters using the above method, the adjustment coefficient factor k usually falls between 0.05 and 0.20. If the k value is too low, it will underestimate the risk brought by the movement trend, resulting in a delayed early warning; if the k value is too high, it will over-predict the risk, which may lead to unnecessary and frequent adjustments to the control system.

[0064] The above calculation formula is further explained as follows: This represents the gradient trend correction term, which is the expected stress growth determined by both the moving rate and the stress gradient. Based on the development trend of the stress concentration area, the base stress value is dynamically corrected.

[0065] The study dynamically predicted the additional stress impact that stress concentration regions might cause at the hard-soft interface during their development. When the stress concentration region rapidly moves towards the hard-soft interface and the stress gradient is large, the gradient trend correction term generates a significant stress increment, thus providing an early warning of potential fracture risks.

[0066] Finally, when the movement trend of the stress concentration area continues to point towards the soft-hard interface area for more than three consecutive acquisition cycles, the theoretical stress peak value is confirmed as the stress peak value.

[0067] It should be noted that a continuous movement trend pointing towards the soft-hard interface for three or more consecutive acquisition cycles can eliminate misjudgments of direction caused by instantaneous fluctuations in the stress concentration area, ensuring the stability of the trend. Only this stable trend can prove that the stress concentration will continue to act on the soft-hard interface, making the calculated theoretical stress peak consistent with the actual stress state of the area.

[0068] The specific steps for locating potential fracture points and assessing risk levels are as follows: Locate the point corresponding to the stress peak as the potential fracture point.

[0069] Based on the mechanical properties of materials, it can be concluded that during the bending process of the rigid-flexible bonded plate, the position corresponding to the stress peak bears the greatest mechanical load. When the stress at this position exceeds the material's own fracture resistance limit, the probability of fracture is much higher than in other areas with lower stress. Therefore, the point corresponding to the stress peak is directly located as the potential fracture point.

[0070] Extract the stress peak value corresponding to the same potential fracture point in the current acquisition cycle and the previous two acquisition cycles, and take the average increase of the stress peak value per unit time as the peak growth rate; then, when the peak growth rate does not exceed the preset growth rate threshold, it is rated as a normal risk level, otherwise it is rated as a high risk level.

[0071] It should be noted that the specific implementation steps for obtaining the preset growth rate threshold are as follows: First, the yield strength of the flexible material in the rigid-flex plate is obtained through material mechanics testing; then, a safety factor is selected according to the process safety requirements, and the ratio of the yield strength to the safety factor is calculated to obtain the allowable stress; next, the ratio of the allowable stress to the standard duration of the entire folding process is recorded as the average stress growth rate threshold; finally, the average stress growth rate threshold is compared with the normal peak growth rate observed in historical qualified process data, and the smaller of the two is taken as the final preset growth rate threshold.

[0072] The stress control module is used to control the bending actuator to adjust the bending radius and speed in real time based on potential fracture points and risk levels, and to construct a stress distribution map.

[0073] See Figure 3As shown, considering the potential fracture points with different spatial distributions and risk levels, their sensitivity to bending radius and speed and the requirements for risk response vary. It is necessary to first determine the control priority based on the spatial distribution characteristics and risk level of the fracture points, and then adjust the bending parameters accordingly. The specific control content is as follows: calculate the distance between each potential fracture point and its nearest neighbor, and take the average of these distances as a reference value for the spatial distribution characteristics of the fracture points.

[0074] When the nearest neighbor distance of a potential fracture point is less than the reference value of the spatial distribution characteristics of fracture points, the fracture point is determined to be located in a dense region; otherwise, the fracture point is determined to be located in a sparse region.

[0075] Potential breakpoints located in densely populated areas and with high-risk levels are identified as priority control areas.

[0076] Potential breakpoints located in sparse regions but with high risk levels are identified as secondary priority control areas.

[0077] Potential breakpoints with conventional risk levels were identified as three-priority control areas.

[0078] Given that a high risk level indicates the urgency of a fracture and is the primary basis for deciding whether intervention is necessary; while the spatial distribution of fracture points reflects the global impact of the risk and the potential speed of structural integrity loss.

[0079] Therefore, if a high-risk point located in a dense area breaks, it is likely to trigger a chain reaction due to the concentration of points, leading to damage to a wider range of structures. It is also more sensitive to changes in bending parameters and is therefore given the highest level of intervention authority. High-risk points in sparse areas may affect local areas, but their spatial mutual influence is small and the risk spread is limited, thus they are given the second level of intervention authority. Conventional risk points, on the other hand, are placed under monitoring rather than immediate intervention because their current threat level is low, thereby avoiding overreaction of the control system and maintaining the stability and efficiency of the process.

[0080] First, a first parameter adjustment strategy is adopted for the control and bending actuator in a priority control area: increase the bending radius and decrease the bending speed.

[0081] Secondly, a second parameter adjustment strategy is adopted for the two-priority control region control bending actuator: the increase in bending radius and the decrease in bending speed are both less than those of the first strategy.

[0082] Finally, for the three-priority control regions, the bending actuator is controlled to maintain the current bending parameters unchanged.

[0083] Specifically, for a priority control area, the high density of risk points indicates that the area is approaching its overall load-bearing limit. Therefore, it is necessary to significantly increase the bending radius to directly reduce the bending curvature and theoretical stress, while simultaneously significantly reducing the bending speed to weaken the dynamic load. For example, if the baseline parameters are a radius of 3 mm and a speed of 10 mm / s, the first parameter adjustment strategy can be implemented by increasing the radius by 50% to 4.5 mm and simultaneously reducing the speed by 40% to 6 mm / s. Through this combination of strong parameters, rapid stress attenuation can be achieved.

[0084] For the second-priority control area, the sparse distribution of risk points indicates that the problem is currently only localized, and there is no need for radical adjustments that could severely impact the overall production cycle. Therefore, the parameter adjustment range should be significantly smaller than the first strategy to mitigate the risk in a more gentle manner. For example, under the same baseline parameters in the first-priority control area, the second parameter adjustment strategy could be specifically manifested as increasing the radius by 20% to 3.6mm while decreasing the speed by 15% to 8.5mm / s. This would effectively alleviate stress at the risk points while minimizing interference with production efficiency.

[0085] The stress distribution map construction steps are as follows: Based on the adjusted stress distribution data, the coordinates of potential fracture points and their corresponding risk levels, a stress distribution cloud map of the rigid-soft bonded plate is generated. The stress distribution cloud map is marked with the distribution of stress magnitude and the location and risk level of potential fracture points. The stress distribution cloud map is updated in real time to reflect the stress changes after adjustment.

[0086] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0087] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0088] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0089] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0090] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A monitoring and control system for the assembly and bending process of a rigid-flex PCB, characterized in that: include: The data acquisition module is used to collect strain rate and stress distribution data of the rigid-flex plate in real time during the bending process; The anomaly comparison module is used to compare strain rate and stress distribution data with standard process data to determine the abnormal change rate. The stress concentration location module is used to determine the density of abnormal points based on the abnormal change rate, locate the stress concentration area based on the density of abnormal points, and determine the stress change gradient and movement trend of the stress concentration area. The fracture risk assessment module is used to calculate the stress peak at the soft-hard interface based on the stress change gradient and movement trend, locate potential fracture points based on the stress peak and assess the risk level. The stress control module is used to control the bending actuator to adjust the bending radius and speed in real time based on potential fracture points and risk levels, and to construct a stress distribution map. The specific contents of the anomaly comparison module are as follows: calculate the deviation of the real-time values ​​of each sensor point in the real-time strain rate data and real-time stress distribution data from the corresponding benchmark values ​​in the standard process data; generate a real-time deviation distribution map based on the deviation of all points, identify the data points located in the high value area in the distribution map, and determine the sensor positions corresponding to these data points as initial anomaly points. The historical deviation data of each initial outlier point within multiple consecutive acquisition periods are traced to form a data sequence. If the data sequence shows a monotonically increasing characteristic within three or more consecutive acquisition periods, it is identified as a valid outlier point. The average change per unit time of each valid outlier point during the trend period is calculated as the outlier change rate of each point.

2. The monitoring and control system for the assembly and folding process of rigid-flex printed circuit boards according to claim 1, characterized in that: The specific contents of the data acquisition module are as follows: Strain sensor groups are arranged on the surface of the rigid-soft bond plate at a preset base spacing, and stress sensors are symmetrically arranged on both sides of the boundary line in the rigid-soft bond area at a spacing smaller than the base spacing. Establish the plate coordinate mapping relationship between all sensors and the motion coordinate system of the bending actuator; The bending actuator is activated to perform bending operations, and the deformation of each strain sensor is read in real time at continuous time points. The local strain rate is calculated based on the deformation changes at adjacent time points. The stress values ​​measured by each stress sensor are read synchronously, and the stress values ​​at each location are fitted into stress distribution data by spatial interpolation based on their position coordinates.

3. The system according to claim 2, wherein the system further comprises a control system. The specific steps for determining the cluster density of outliers based on the rate of abnormal change are as follows: The rigid-soft board is divided into multiple analysis units. The number of valid outliers falling into each analysis unit is counted, and the arithmetic mean of the abnormal change rate of all valid outliers in each analysis unit is calculated. The outlier cluster density index for each analysis unit is generated by multiplying the number of valid outliers by the average rate of change of outliers, and is recorded as the outlier cluster density of each analysis unit.

4. The system according to claim 1, wherein the system further comprises a control system. The specific steps for locating stress concentration areas based on the density of abnormal points are as follows: Analysis units with anomaly density exceeding a preset density threshold are identified as candidate units, and adjacent candidate units are determined to be adjacent to each other when they share a boundary or corner point. Adjacent candidate elements are merged to form a continuous initial stress concentration region; When the density of anomalous points in each preliminary stress concentration area does not decrease within three or more consecutive acquisition cycles, the area is confirmed as an effective stress concentration area. Identify all the peripheral analysis units that constitute the effective stress concentration region, extract the coordinates of the outer boundary vertices of each peripheral analysis unit, connect the outer boundary vertices in sequence to form a closed boundary polygon, and record the area enclosed by the boundary polygon as the boundary range of the stress concentration region. Output the boundary coordinates and spatial location information of all effective stress concentration regions.

5. The system according to claim 1, wherein: the system further comprises a control system. The steps for determining the stress change gradient and movement trend are as follows: Determine the stress maximum and minimum points within the stress concentration region, calculate the difference vector between the coordinates of the two points, record the direction of the difference vector as the stress change gradient direction, and calculate the ratio of the stress difference between the two points to their distance to obtain the stress change gradient value. Extract the coordinates of the stress maximum points in the current cycle and the previous two cycles. Take the vector direction from the first cycle coordinates to the current cycle coordinates as the movement direction, obtain the coordinate distance between the current cycle and the first cycle, and calculate the ratio with the corresponding time interval to obtain the movement rate. The direction of movement and the rate of movement are used together to determine the stress movement trend in the stress concentration area.

6. The monitoring and control system for the flexible-rigid bonding board assembly bending process according to claim 1, characterized in that: The calculation process for the peak stress at the soft-hard interface is as follows: Draw an extension line along the stress gradient from the point of maximum stress to the soft-hard interface region, and determine the location of the intersection of the extension line and the soft-hard interface region. When the intersection point is located within the soft-hard boundary zone, and the direction of movement of the stress concentration area points towards the soft-hard boundary zone, the stress concentration area is determined to constitute a trend influence. The weighted sum of the maximum stress value and the stress change gradient value in the stress concentration area that constitutes the trend influence is taken as the theoretical stress peak at the soft-hard interface, where the moving rate is used as the weighting factor. When the movement trend of the stress concentration area continues to point towards the soft-hard interface area for more than three consecutive acquisition cycles, the theoretical stress peak value is confirmed as the stress peak value.

7. The system according to claim 1, wherein: the system further comprises a control system. The specific steps for locating potential breakpoints and assessing risk levels are as follows: The point corresponding to the stress peak is identified as the potential fracture point; Extract the stress peak value corresponding to the same potential fracture point in the current acquisition cycle and the previous two acquisition cycles, and take the average increase of the stress peak value per unit time as the peak growth rate. When the peak growth rate does not exceed the preset growth rate threshold, it is assessed as a normal risk level; otherwise, it is assessed as a high risk level.

8. The system according to claim 1, wherein: the system further comprises a control system. The specific details of controlling the bending actuator to adjust the bending radius and speed in real time based on potential breakpoints and risk levels are as follows: Calculate the distance between each potential fracture point and its nearest neighbor, and take the average of these distances as a reference value for the spatial distribution characteristics of fracture points; When the nearest neighbor distance of a potential break point is less than the reference value of the spatial distribution characteristics of the break point, the break point is determined to be located in a dense region; otherwise, the break point is determined to be located in a sparse region. Potential breakpoints located in densely populated areas and with high-risk levels are identified as priority control areas. Potential breakpoints located in sparse regions but with high-risk levels are identified as second-priority control areas. Potential breakpoints with conventional risk levels are identified as three-priority control zones. For a priority control region, the first parameter adjustment strategy for the bend control actuator is adopted: increase the bend radius and decrease the bend speed; For the two-priority control region control bypass actuator, a second parameter adjustment strategy is adopted: the increase in bypass radius and the decrease in bypass speed are both less than the first strategy; For the three-priority control regions, the control mechanism maintains the current bending parameters unchanged.

9. A monitoring and control system for the flexible-rigid bonded board assembly bending process according to claim 1, characterized in that: The steps for constructing the stress distribution map are as follows: Based on the adjusted stress distribution data, the coordinates of potential fracture points, and the corresponding risk levels, a stress distribution cloud map of the rigid-soft bonded plate is generated. The stress distribution is marked in the stress distribution cloud map, and the location of potential fracture points and their risk levels are also marked.

Images