A method and system for controlling the pressing of a multilayer rigid-flex circuit board

Abstract

The present application relates to a kind of multilayer rigid-flex combination circuit board pressing control method and system based on, it relates to data processing field, including: a kind of multilayer rigid-flex combination circuit board pressing control method and system are provided, by utilizing hierarchical clustering partition technology, according to the partition of pressure and temperature according to pressing plate pressing log, the pressure and temperature distribution situation of accurate identification pressing plate in different regions can be identified;Temperature field simulation is carried out by the time series information of pressure and temperature of target partition, the temperature distribution situation of multilayer rigid-flex circuit board can be monitored and predicted in real time, when temperature field time series information is consistent with standard, by the time series information of pressure and temperature control pressing plate executes pressing control, the pressing quality of multilayer rigid-flex combination circuit board is significantly improved, to effectively solve the technical problems such as the great temperature and pressure fluctuation and uneven distribution caused by the difference of material thermal conductivity, multilayer interface heat transfer lag and pressing plate wear etc.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a method and system for controlling the lamination of multilayer rigid-flex circuit boards. Background Technology

[0002] In the field of electronics manufacturing, multilayer rigid-flex circuit boards are widely used in various complex electronic devices due to their combination of the stability of rigid boards and the flexibility of flexible boards. In their manufacturing process, lamination is one of the key steps. Pressure and temperature are applied to the multilayer circuit board using a laminator to achieve interlayer bonding. Traditional lamination control methods mainly rely on preset temperature and pressure parameters, which are usually set based on experience and therefore fluctuate significantly during the lamination process.

[0003] On the one hand, the thermal conductivity of rigid and flexible materials differs significantly, leading to varying heat transfer rates between different material layers. On the other hand, after multilayer boards are stacked, heat needs to penetrate more interfaces, causing the inner layer temperature to lag behind the outer layer. These factors result in significant temperature and pressure fluctuations. Furthermore, during long-term use, the laminating platen may experience uneven pressure and temperature distribution due to wear, further increasing the difficulty of temperature and pressure control for multilayer rigid-flex circuit boards. Currently, the market lacks sophisticated temperature and pressure control solutions for laminating multilayer rigid-flex circuit boards, making it difficult to meet the demands of high-quality lamination.

[0004] Current lamination control methods mostly rely on manual experience to set temperature and control schemes, lacking precise and effective control schemes. This results in technical problems such as imprecise temperature and pressure control during the lamination of multilayer rigid-flex circuit boards, which affects the lamination performance of multilayer rigid-flex circuit boards. Summary of the Invention

[0005] This invention addresses the technical problem in the prior art where the temperature and pressure control during the lamination of multilayer rigid-flex circuit boards is not precise, thus affecting the lamination performance of multilayer rigid-flex circuit boards. It provides a method and system for controlling the lamination of multilayer rigid-flex circuit boards to solve this problem.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0007] In a first aspect, the present invention provides a method for controlling the lamination of multilayer rigid-flex circuit boards, comprising:

[0008] Based on the pressure plate pressing log, pressure plate pressure hierarchical clustering and partitioning are performed to obtain pressure plate partitioning;

[0009] Based on the pressing log, the pressing temperature is hierarchically clustered and partitioned to obtain the pressing temperature partition;

[0010] The intersection of the pressure zone and the temperature zone of the pressure plate is taken to obtain the target zone of the pressure plate.

[0011] Receive pressure timing information and temperature timing information, perform frequent temperature and pressure statistics according to the target partition of the pressure plate, and obtain the target partition pressure timing information and target partition temperature timing information;

[0012] Based on the target partition pressure timing information and the target partition temperature timing information, temperature field simulation is performed on the geometry of the multilayer rigid-flex circuit board to obtain temperature field timing information.

[0013] When the temperature field timing information is consistent with the standard temperature field timing information, the pressure plate is controlled to perform pressing control according to the pressure timing information and the temperature timing information.

[0014] Secondly, the present invention provides a multilayer rigid-flex circuit board lamination control system, comprising:

[0015] The pressure plate partitioning template is used to perform hierarchical clustering and partitioning of pressure plate based on the pressure plate pressing log to obtain the pressure plate partitioning.

[0016] The platen temperature partitioning module is used to perform hierarchical clustering and partitioning of platen temperature based on the platen pressing log to obtain platen temperature partitions.

[0017] The pressure plate target partitioning module is used to take the intersection of the pressure plate pressure partitioning and the pressure plate temperature partitioning to obtain the pressure plate target partitioning;

[0018] The pressure and temperature timing information module is used to receive pressure timing information and temperature timing information, and perform frequent temperature and pressure statistics according to the target partition of the pressure plate to obtain the target partition pressure timing information and the target partition temperature timing information.

[0019] The temperature field timing information module is used to simulate the temperature field of the multilayer rigid-flex circuit board geometry based on the target partition pressure timing information and the target partition temperature timing information to obtain temperature field timing information.

[0020] The pressing control module is used to control the pressure plate to perform pressing control according to the pressure timing information and the temperature timing information when the temperature field timing information is consistent with the standard temperature field timing information.

[0021] The beneficial effects of this invention are as follows: This invention provides a method and system for controlling the lamination of multilayer rigid-flex circuit boards. Compared with traditional lamination control that relies on manual experience to set temperature and control schemes, this invention utilizes hierarchical clustering partitioning technology to partition pressure and temperature according to the lamination log, enabling accurate identification of pressure and temperature distribution in different areas of the lamination board and effectively solving the problem of uneven distribution caused by lamination wear. By taking the intersection to obtain the target partition of the lamination board, and combining pressure and temperature time series information for frequent temperature and pressure statistics, the accuracy of temperature and pressure control is further refined. By simulating the temperature field using the pressure and temperature time series information of the target partition, the temperature distribution of the multilayer rigid-flex circuit board can be monitored and predicted in real time, and compared with standard temperature field time series information to ensure that the temperature field during the lamination process meets expectations. When the temperature field timing information is consistent with the standard, the pressure plate is controlled to perform pressing control through pressure and temperature timing information. This effectively solves the problems of large temperature and pressure fluctuations and uneven distribution caused by differences in material thermal conductivity, lag in heat transfer at multi-layer interfaces, and pressure plate wear. It significantly improves the pressing quality of multi-layer rigid-flex circuit boards and solves the technical problem of precise temperature and pressure control in the pressing of multi-layer rigid-flex circuit boards. Attached Figure Description

[0022] Figure 1 A flowchart illustrating a multilayer rigid-flex circuit board lamination control method provided by the present invention;

[0023] Figure 2 This is a schematic diagram of a multilayer rigid-flex circuit board lamination control system provided by the present invention.

[0024] Reference numerals: 11. Pressure plate pressure partitioning template; 12. Pressure plate temperature partitioning module; 13. Pressure plate target partitioning module; 14. Pressure and temperature timing information module; 15. Temperature field timing information module; 16. Pressing control module. Detailed Implementation

[0025] 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.

[0026] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0028] Example 1:

[0029] like Figure 1 As shown, this embodiment of the invention provides a method for controlling the lamination of multilayer rigid-flex circuit boards, which specifically includes the following steps:

[0030] S10: Perform pressure hierarchical clustering and partitioning based on the pressure plate pressing log to obtain the pressure plate partitioning;

[0031] Furthermore, based on the pressure plate pressing log, pressure plate pressure hierarchical clustering and partitioning are performed to obtain pressure plate partitioning. Step S10 includes:

[0032] S11: Obtain the starting geometry and target geometry of the multilayer rigid-flex circuit board geometry;

[0033] Furthermore, the press plate lamination log refers to a file that records various parameters and states of the press plate during the lamination of multilayer rigid-flex circuit boards, including information such as time, pressure, temperature, and location; hierarchical clustering partitioning refers to a data processing method that groups data points into clusters with similar characteristics by calculating the similarity or distance between data points and organizes them according to a hierarchical structure; the initial geometry and target geometry refer to the initial shape and size of the circuit board before lamination, and the target geometry refers to the final shape and size of the circuit board expected to achieve after lamination; temperature time-series recording information refers to data recording the temperature change over time during the lamination process; and pressure distribution recording information refers to recording the pressure distribution of the press plate in different areas during the lamination process.

[0034] S12: Retrieve the pressure plate pressing log of the target position plate that satisfies the starting geometry and the target geometry;

[0035] S13: Extract the first pressure distribution record information up to the Nth pressure distribution record information from the pressure plate pressing log, where the temperature timing record information is the same;

[0036] S14: Based on the first pressure distribution record information up to the Nth pressure distribution record information, perform pressure plate pressure hierarchical clustering and partitioning to obtain pressure plate pressure partitions.

[0037] For example, the starting geometry and target geometry of the multilayer rigid-flex circuit board are obtained: it is assumed that the starting geometry of the multilayer rigid-flex circuit board is a multilayer composite with a thickness of 1.5mm, containing 3 rigid layers and 2 flexible layers, and the target geometry is a flat circuit board with a thickness of 1.2mm, with each layer tightly bonded.

[0038] The pressing logs of the target position platen that satisfy the initial geometry and the target geometry are retrieved, and the historical pressing logs of the target position platen are retrieved from the pressing equipment's database. Assume the retrieved pressing logs record the following data:

[0039] time Temperature (°C) Pressure (MPa) Position (x, y) 0s 150 0.5 (0,0) 10s 150 0.6 (10,10) 20s 150 0.7 (20,20) 30s 150 0.8 (30,30) 40s 150 0.9 (40,40) 50s 150 1.0 (50,50)

[0040] From the pressure plate pressing log, extract the first pressure distribution record information up to the Nth pressure distribution record information with the same temperature time sequence information. Assuming the temperature time sequence information is 150℃, extract the corresponding pressure distribution record information:

[0041] time Pressure (MPa) Position (x, y) 0s 0.5 (0,0) 10s 0.6 (10,10) 20s 0.7 (20,20) 30s 0.8 (30,30) 40s 0.9 (40,40) 50s 1.0 (50,50)

[0042] Based on the first pressure distribution record information up to the Nth pressure distribution record information, hierarchical clustering and partitioning of the pressure plate are performed to obtain pressure plate partitions. A hierarchical clustering algorithm is used to partition the extracted pressure distribution record information. Assuming Euclidean distance is used as the similarity metric, the distance matrix between each point is calculated:

[0043] coordinates (0,0) (10,10) (20,20) (30,30) (40,40) (50,50) (0,0) 0 14.14 28.28 42.43 56.57 70.71 (10,10) 14.14 0 14.14 28.28 42.43 56.57 (20,20) 28.28 14.14 0 14.14 28.28 42.43 (30,30) 42.43 28.28 14.14 0 14.14 28.28 (40,40) 56.57 42.43 28.28 14.14 0 14.14 (50,50) 70.71 56.57 42.43 28.28 14.14 0

[0044] Based on the hierarchical clustering algorithm, these points are divided into multiple partitions. Assume the final partitioning consists of 3 partitions:

[0045] Partition 1: (0,0), (10,10)

[0046] Partition 2: (20,20), (30,30)

[0047] Partition 3: (40,40), (50,50)

[0048] Further, based on the first pressure distribution record information up to the Nth pressure distribution record information, pressure plate pressure hierarchical clustering and partitioning are performed to obtain pressure plate pressure partitions. Step S14 includes:

[0049] S141: Perform neighborhood hierarchical clustering analysis on the first pressure distribution record information to obtain the first pressure distribution partition;

[0050] S142: Until neighborhood hierarchical clustering analysis is performed on the Nth pressure distribution record information to obtain the Nth pressure distribution partition;

[0051] S143: Perform pairwise enumeration and evaluation of the distribution similarity of the first pressure distribution partition up to the Nth pressure distribution partition to obtain several partition similarities;

[0052] S144: Based on the similarity of the several partitions, outlier distribution partitions are deleted from the first pressure distribution partition up to the Nth pressure distribution partition to obtain a concentrated pressure distribution partition. The intersection of the concentrated pressure distribution partitions is taken as the pressure plate partition.

[0053] For example, neighborhood hierarchical clustering analysis is performed on the first pressure distribution record information to obtain the first pressure distribution partition. Assume the first pressure distribution record information is as follows:

[0054] time Pressure (MPa) Position (x, y) 0s 0.5 (0,0) 10s 0.6 (10,10) 20s 0.7 (20,20)

[0055] Neighborhood hierarchical clustering analysis was used to calculate the similarity of each data point within its local neighborhood. Assuming Euclidean distance was used as the similarity metric, the results are as follows:

[0056] The distance between (0,0) and (10,10) is 14.14.

[0057] The distance between (10,10) and (20,20) is 14.14.

[0058] Based on the hierarchical clustering algorithm, these points are divided into two partitions:

[0059] Partition 1.1: (0,0), (10,10)

[0060] Partition 1.2: (20,20)

[0061] Perform neighborhood hierarchical clustering analysis on the Nth pressure distribution record information to obtain the Nth pressure distribution partition. Assume the Nth pressure distribution record information is as follows:

[0062] time Pressure (MPa) Position (x, y) 30s 0.8 (30,30) 40s 0.9 (40,40) 50s 1.0 (50,50)

[0063] Neighborhood hierarchical clustering analysis was used to calculate the similarity of each data point within its local neighborhood. Assuming Euclidean distance was used as the similarity metric, the results are as follows:

[0064] The distance between (30,30) and (40,40) is 14.14.

[0065] The distance between (40,40) and (50,50) is 14.14.

[0066] Based on the hierarchical clustering algorithm, these points are divided into two partitions:

[0067] Partition N.1: (30,30), (40,40)

[0068] Partition N.2: (50, 50)

[0069] The distribution similarity of the first pressure distribution partition up to the Nth pressure distribution partition is evaluated pairwise to obtain several partition similarities. The similarity between partition 1.1 and partition N.1 is calculated, assuming that the average pressure value is used as the similarity evaluation standard:

[0070] The average pressure value for zone 1.1 is 0.55.

[0071] The average pressure value for zone N.1 is 0.85.

[0072] Calculate similarity:

[0073] Similarity = |0.55 - 0.85| = 0.3

[0074] Similarly, calculate the similarity between other partitions:

[0075] The similarity between partition 1.2 and partition N.2 is |0.7 - 1.0| = 0.3

[0076] Based on the similarity of the aforementioned partitions, outlier partitions are removed from the first pressure distribution partition up to the Nth pressure distribution partition to obtain concentrated pressure distribution partitions. The intersection of these concentrated pressure distribution partitions is then taken as the pressure plate partition.

[0077] Based on the similarity assessment results, partitions with a similarity greater than 0.2 are considered outliers and need to be deleted. Therefore:

[0078] The similarity between partition 1.1 and partition N.1 is 0.3, so partition 1.1 and partition N.1 should be deleted.

[0079] The similarity between partition 1.2 and partition N.2 is 0.3, so partitions 1.2 and N.2 should be deleted.

[0080] After filtering, the remaining partitions are:

[0081] Centralized pressure distribution zoning: None

[0082] Since all partitions have been deleted, it means that no partitions with similar stress distribution characteristics can be found in the current data. In this case, the similarity threshold can be adjusted, for example, by increasing the threshold to 0.4, and the filtering can be performed again.

[0083] Assuming that after adjustment, the similarity between partitions 1.2 and N.2 meets the requirements, they are retained. The intersection of these two partitions is taken as the pressure partition of the pressure plate:

[0084] Pressure zones on the pressure plate: (20,20), (50,50)

[0085] Furthermore, the distribution similarity of the first pressure distribution partition up to the Nth pressure distribution partition is evaluated pairwise to obtain several partition similarities. Step S143 includes:

[0086] S1431: Connect the centers of the first pressure distribution zones to obtain the first pattern;

[0087] S1432: Connect the centers of the second pressure distribution zones to obtain the second pattern;

[0088] S1433: Calculate the similarity between the first pattern and the second pattern to obtain the first partition similarity, and add it to the plurality of partition similarities.

[0089] For example, the center lines of the first pressure distribution partitions are connected to obtain a first pattern. Assume the first pressure distribution partitions are as follows:

[0090] Partition 1.1: (0,0), (10,10)

[0091] Partition 1.2: (20,20)

[0092] Calculate the center position of each partition:

[0093] The center position of partition 1.1 is: ((0+10) / 2,(0+10) / 2)=(5,5)

[0094] The center of partition 1.2 is located at (20, 20).

[0095] Connect these center points to form the first pattern. Assume the first pattern is a simple line segment connecting points (5,5) and (20,20).

[0096] Connecting the centers of the second pressure distribution zones yields the second pattern. Assume the second pressure distribution zones are as follows:

[0097] Partition N.1: (30,30), (40,40)

[0098] Partition N.2: (50, 50)

[0099] Calculate the center position of each partition:

[0100] The center position of partition N.1 is: ((30+40) / 2,(30+40) / 2)=(35,35)

[0101] The center of partition N.2 is located at (50, 50).

[0102] Connect these center points to form the second pattern. Assume the second pattern is a simple line segment connecting points (35, 35) and (50, 50).

[0103] Calculate the similarity between the first pattern and the second pattern to obtain the first partition similarity, and add it to the plurality of partition similarities.

[0104] To calculate the similarity between two patterns, shape matching algorithms can be used. Let's assume a simple contour similarity calculation method is employed, which assesses similarity by comparing geometric features such as the length and angles of the two patterns.

[0105] The length of the first pattern: the distance from (5,5) to (20,20) is

[0106] The length of the second pattern: the distance from (35,35) to (50,50) is

[0107] Two patterns are the same length, but their positions are different. Assume the similarity calculation formula is: Similarity = 1 / (1 + Distance Difference)

[0108] Calculate the distance between the center points of the two patterns:

[0109] The center point of the first pattern is: ((5+20) / 2,(5+20) / 2)=(12.5,12.5)

[0110] The center point of the second pattern is: ((35+50) / 2,(35+50) / 2)=(42.5,42.5)

[0111] The distance between the center points is:

[0112] Calculate similarity:

[0113] Add this similarity score to the similarity scores of several partitions.

[0114] Through the above embodiments, the pressure plate is divided into multiple pressure zones, and zones with similar pressure distribution characteristics are selected by calculating the pattern similarity of different zones. This method can effectively identify the pressure change patterns of the pressure plate in different areas, especially when the pressure plate wear causes uneven pressure distribution, and can accurately locate the problem area. The resulting pressure zones can provide a refined partitioning basis for subsequent lamination control, thereby effectively solving the problem of uneven pressure distribution caused by pressure plate wear, and achieving the technical effect of improving the lamination quality of multilayer rigid-flex circuit boards.

[0115] S20: Perform hierarchical clustering and partitioning of the platen temperature based on the platen pressing log to obtain the platen temperature partition;

[0116] Furthermore, obtain the pressure plate temperature hierarchical clustering partition and obtain the pressure plate pressure hierarchical clustering partition, and execute step S10.

[0117] S30: Take the intersection of the pressure zone and the temperature zone of the pressure plate to obtain the target zone of the pressure plate;

[0118] Further, step S143 is executed to obtain the target partition of the pressure plate.

[0119] S40: Receive pressure timing information and temperature timing information, perform frequent temperature and pressure statistics based on the target partition of the pressure plate, obtain the target partition pressure timing information and the target partition temperature timing information, and execute step S40, which includes:

[0120] S41: Based on the multilayer rigid-flex circuit board model and the service life of the laminator, collect the first partition temperature timing information set up to the Mth partition temperature timing information set, and the first partition pressure timing information set up to the Mth partition pressure timing information set, which satisfy the pressure timing information and the temperature timing information.

[0121] S42: Simultaneous ensemble value evaluation is performed on the temperature time series information set of the first partition up to the temperature time series information set of the Mth partition, and the pressure time series information set of the first partition up to the pressure time series information set of the Mth partition, respectively, to obtain the pressure time series information and temperature time series information of the target partition.

[0122] For example, the temperature timing information set of the first partition up to the Mth partition temperature timing information set, and the pressure timing information set of the first partition up to the Mth partition pressure timing information set are collected, assuming the following constraints on the multilayer rigid-flex circuit board model and the service life of the laminator:

[0123] Circuit board model: Model A

[0124] Press machine service life: 2 years

[0125] Collect pressure and temperature timing information that meets the above conditions from the database of the pressing equipment. Assume the collected data is as follows:

[0126] partition Time (s) Pressure (MPa) Temperature (°C) 1.1 0 0.5 150 1.1 10 0.6 150 1.1 20 0.7 150 1.2 0 0.8 150 1.2 10 0.9 150 1.2 20 1.0 150 N.1 0 0.6 150 N.1 10 0.7 150 N.1 20 0.8 150 N.2 0 0.9 150 N.2 10 1.0 150 N.2 20 1.1 150

[0127] These data are classified into the first partition temperature time series information set, the first partition pressure time series information set, and so on, up to the Mth partition temperature time series information set and the Mth partition pressure time series information set.

[0128] Simultaneous moment-based value evaluation is performed on the temperature time-series information set of the first partition up to the temperature time-series information set of the Mth partition, and the pressure time-series information set of the first partition up to the pressure time-series information set of the Mth partition, respectively, to obtain the pressure time-series information and temperature time-series information of the target partition.

[0129] Simultaneous central tendency evaluation is performed on the time-series temperature and pressure information for each partition. Taking partitions 1.1 and N.1 as examples, the central tendency (e.g., average value) is calculated for each time point.

[0130]

[0131] Calculate the concentrated values ​​at the same time:

[0132] Time 0s:

[0133] Average pressure: (0.5 + 0.6) / 2 = 0.55 MPa

[0134] Average temperature: (150 + 150) / 2 = 150℃

[0135] Time: 10 seconds

[0136] Average pressure: (0.6 + 0.7) / 2 = 0.65 MPa

[0137] Average temperature: (150 + 150) / 2 = 150℃

[0138] Time: 20s

[0139] Average pressure: (0.7 + 0.8) / 2 = 0.75 MPa

[0140] Average temperature: (150 + 150) / 2 = 150℃

[0141] These concentrated values ​​are used as the pressure and temperature time series information for the target partition.

[0142] Through the above steps, refined temperature and pressure statistics for the target zone of the pressure plate are achieved. By collecting pressure and temperature timing information that meets specific conditions and performing simultaneous lumped value evaluation, the pressure and temperature timing information of the target zone can be accurately obtained. This information provides important data support for subsequent lamination control, which helps to achieve refined temperature control in the lamination of multilayer rigid-flex circuit boards. This effectively solves the problems of large temperature and pressure fluctuations and uneven distribution caused by differences in material thermal conductivity, lag in heat transfer at multilayer interfaces, and pressure plate wear, and significantly improves the lamination quality of multilayer rigid-flex circuit boards.

[0143] S50: Based on the target partition pressure timing information and the target partition temperature timing information, perform temperature field simulation on the geometry of the multilayer rigid-flex circuit board to obtain temperature field timing information.

[0144] Furthermore, based on the target partition pressure timing information and the target partition temperature timing information, a temperature field simulation is performed on the geometry of the multilayer rigid-flex circuit board to obtain temperature field timing information. Step S50 includes:

[0145] S51: Based on the target partition of the pressure plate, construct the pressure plate mesh topology, combine it with the multi-layer rigid-flex circuit board mesh, perform topological twinning based on graph neural network, and build a geometric temperature field fitting model architecture. The input node of the geometric temperature field fitting model architecture is the target partition of the pressure plate, and the input data of any partition is the pressure time series information and temperature time series information. The output node is the mesh temperature field of the multi-layer rigid-flex circuit board mesh. The multi-layer rigid-flex circuit board mesh will change from the initial mesh to the target mesh according to the preset shrinkage speed.

[0146] S52: Using the multilayer rigid-flex circuit board model and the laminator model as constraints, collect pressure time-series record information, temperature time-series record information and temperature field time-series record information of the first multilayer rigid-flex circuit board geometry with a service duration less than or equal to the service duration threshold, train the geometry temperature field fitting model architecture, and obtain the temperature field fitting base model.

[0147] S53: Using the multilayer rigid-flex circuit board model and laminating machine location as constraints, collect the target partition pressure timing record information, the target partition temperature timing record information, and the second multilayer rigid-flex circuit board geometry temperature field timing record information, train the temperature field fitting basis model, obtain the temperature field fitting model, and perform temperature field simulation on the multilayer rigid-flex circuit board geometry based on the target partition pressure timing information and the target partition temperature timing information to obtain temperature field timing information.

[0148] For example, a pressure plate mesh topology is constructed, combined with a multilayer rigid-flex circuit board mesh, and a topological twin is generated based on a graph neural network to build a geometric temperature field fitting model architecture. The mesh topologies of the pressure plate and the multilayer rigid-flex circuit board are assumed to be as follows:

[0149] Pressure plate mesh topology: The pressure plate is divided into 4 partitions, each containing several mesh cells.

[0150] Multilayer rigid-flex circuit board mesh: The circuit board is divided into several mesh units, and each mesh unit corresponds to a temperature value.

[0151] Topological twinning is performed based on graph neural networks, mapping the pressure plate mesh topology to the circuit board mesh volume to form a topological twin relationship. The model architecture is as follows:

[0152] Input nodes: target partitions of the pressure plate. The input data for each partition are pressure timing information and temperature timing information.

[0153] Output node: Temperature field distribution of multilayer rigid-flex circuit board mesh.

[0154] Constrained by the model of the multilayer rigid-flex circuit board and the model of the laminator, pressure time-series records, temperature time-series records, and temperature field time-series records of the geometry of the first multilayer rigid-flex circuit board are collected when the service time is less than or equal to the service time threshold. The geometry temperature field fitting model architecture is trained to obtain the temperature field fitting base model. Assume the collected data is as follows:

[0155] Circuit board model: Model A

[0156] Press machine model: Model B

[0157] Service duration threshold: 2 years

[0158] The collected training data is as follows:

[0159]

[0160] Use these data to train the geometry temperature field fitting model architecture and obtain the temperature field fitting base model.

[0161] Using the multilayer rigid-flex circuit board model and laminating machine location number as constraints, the target partition pressure time-series record information, the target partition temperature time-series record information, and the second multilayer rigid-flex circuit board geometry temperature field time-series record information are collected. The temperature field fitting basis model is trained to obtain the temperature field fitting model. Assume the collected target partition data is as follows:

[0162]

[0163]

[0164] Use these data to train the temperature field fitting basis model to obtain the temperature field fitting model.

[0165] Furthermore, the data model training specifically involves:

[0166] Initialize model parameters. Before training begins, the model parameters need to be initialized. The initialization method varies depending on the model. For example, in neural networks, small random values ​​are typically used to initialize the weights to break symmetry and allow the model to learn normally.

[0167] Define the loss function. The loss function measures the difference between the model's predictions and the true values. For classification problems, the commonly used loss function is the cross-entropy loss function; for regression problems, the commonly used loss function is the mean squared error loss function. For example, in a binary classification problem, the cross-entropy loss function can be expressed as:

[0168]

[0169] Where y is the real label. N is the probability predicted by the model, and N is the number of samples.

[0170] Choosing an optimization algorithm. Optimization algorithms are used to adjust the model's parameters based on the loss function to minimize the loss value. Common optimization algorithms include gradient descent, stochastic gradient descent (SGD), mini-batch gradient descent, and the Adam algorithm.

[0171] Taking stochastic gradient descent as an example, it calculates the gradient using only one sample at a time and then updates the model parameters. This method is fast, but it may cause large fluctuations in the loss value during training. Mini-batch gradient descent, on the other hand, calculates the gradient using a batch of samples each time, thus balancing computational speed and stability.

[0172] Iterative training. During training, the model iterates continuously, updating its parameters with each iteration based on the optimization algorithm. In each iteration, the model receives input data, calculates the predicted value through forward propagation, calculates the loss value using the loss function, and finally calculates the gradient and updates the parameters through backpropagation.

[0173] The number of iterations is typically determined by hyperparameters (such as the maximum number of iterations) or convergence criteria (such as the loss value no longer decreasing significantly). For example, training a simple linear regression model may only require a few dozen iterations to converge; while for complex deep learning models, it may require thousands or even tens of thousands of iterations.

[0174] Furthermore, based on the target partition pressure timing information and the target partition temperature timing information, a temperature field simulation is performed on the geometry of the multilayer rigid-flex circuit board to obtain temperature field timing information. Assume the input target partition pressure and temperature timing information are as follows:

[0175]

[0176] Simulations were performed using a pre-trained temperature field fitting model to obtain time-series information about the temperature field.

[0177] Time (s) Circuit board temperature field (°C) 0 [150,150,150] 10 [150,150,150] 20 [150,150,150]

[0178] Through the above steps, the temperature field simulation of the geometry of a multilayer rigid-flex circuit board (FPCB) was achieved. By constructing a topological twin relationship between the pressure plate mesh and the FPCB mesh, and using a graph neural network to train and simulate the temperature field fitting model, the temporal information of the temperature field of the FPCB can be accurately obtained. This information provides important data support for subsequent lamination control, helping to achieve refined temperature control during FPCB lamination. This effectively solves the problems of large temperature and pressure fluctuations and uneven distribution caused by differences in material thermal conductivity, lag in heat transfer at multilayer interfaces, and pressure plate wear, significantly improving the lamination quality of FPCBs.

[0179] S60: When the temperature field timing information is consistent with the standard temperature field timing information, the pressure plate is controlled to perform pressing control according to the pressure timing information and the temperature timing information.

[0180] Furthermore, when the temperature field time series information is inconsistent with the standard temperature field time series information, the pressure time series information and the temperature time series information are updated, and the updated pressure time series information and the updated temperature time series information are used to perform cyclic analysis. At the same time, the pressure time series information and the temperature time series information are stored as the temperature field intervention medium. Step S60 includes:

[0181] S61: When the number of temperature field intervention media is greater than or equal to the threshold number of intervention media, the first, second, and third temperature field intervention media with fitness values ​​from smallest to largest are obtained by using the deviation parameter between the temperature field intervention media and the standard temperature field intervention media as the fitness function.

[0182] S62: With the goal of simultaneously reducing the three deviation parameters of the first temperature field intervention medium, the second temperature field intervention medium, and the third temperature field intervention medium, update other temperature field intervention media to obtain expanded pressure time series information and expanded temperature time series information and perform cyclic analysis.

[0183] For example, when the temperature field time series information is inconsistent with the standard temperature field time series information, the pressure time series information and temperature time series information are updated. The updated pressure time series information and updated temperature time series information are then used to perform a cyclical analysis. Simultaneously, the pressure time series information and temperature time series information are stored as the temperature field intervention medium. Assume the standard temperature field time series information is as follows:

[0184] Time (s) Standard temperature field (°C) 0 [150,150,150] 10 [150,150,150] 20 [150,150,150]

[0185] The simulated temperature field time series information is as follows:

[0186] Time (s) Simulated temperature field (°C) 0 [148,148,148] 10 [149,149,149] 20 [151,151,151]

[0187] Because the simulated temperature field differs from the standard temperature field, the pressure and temperature time series information needs to be updated. Assume the updated pressure and temperature time series information is as follows:

[0188] Time (s) Update pressure (MPa) Update temperature (°C) 0 0.56 151 10 0.66 151 20 0.76 151

[0189] The temperature field simulation was re-performed using the updated information to obtain new temperature field time series information:

[0190] Time (s) Updated temperature field (°C) 0 [149,149,149] 10 [150,150,150] 20 [152,152,152]

[0191] The updated pressure and temperature time series information is stored as a temperature field intervention medium.

[0192] When the number of temperature field intervention media is greater than or equal to the threshold number of intervention media, the deviation parameter between the temperature field intervention media and the standard temperature field intervention media is used as the fitness function to obtain the first, second, and third temperature field intervention media with fitness ranging from small to large. Assuming the threshold number of intervention media is 3, the stored temperature field intervention media are as follows:

[0193]

[0194] The fitness of each intervention medium is calculated using the deviation parameter between the temperature field intervention medium and the standard temperature field intervention medium as the fitness function. The top three intervention media are selected based on their fitness, sorted from smallest to largest.

[0195] First temperature field interference medium: No. 2, deviation parameter 1;

[0196] Second temperature field interference medium: No. 1, deviation parameter 2;

[0197] The third temperature field intervention medium: No. 3, deviation parameter 3.

[0198] With the goal of simultaneously reducing the three deviation parameters of the first, second, and third temperature field intervention media, other temperature field intervention media are updated. Expanded pressure and temperature time series information are obtained, and cyclical analysis is performed. It is assumed that the temperature field intervention medium numbered 4 needs to be updated. Based on the deviation parameter adjustment strategies of the first three intervention media, the intervention medium numbered 4 is updated as follows:

[0199]

[0200]

[0201] The temperature field simulation was re-performed using the updated pressure and temperature time series information to obtain the expanded temperature field time series information:

[0202] Time (s) Expanded temperature field (°C) 0 [149,149,149] 10 [150,150,150] 20 [151,151,151]

[0203] Through the above steps, the temperature field intervention medium is updated and optimized. By storing the pressure and temperature timing information of each simulation and evaluating it using the deviation parameter as a fitness function, the pressure and temperature timing information can be gradually adjusted to make the simulated temperature field closer to the standard temperature field. This method can effectively solve the temperature field deviation problem caused by inaccurate initial parameters or environmental changes, further improve the temperature control accuracy of multilayer rigid-flex circuit board lamination, and ensure lamination quality.

[0204] Example 2:

[0205] like Figure 2 As shown, based on the same inventive concept as the multilayer rigid-flex circuit board lamination control method provided in Embodiment 1, this embodiment of the invention also provides a multilayer rigid-flex circuit board lamination control system. The explanation of the multilayer rigid-flex circuit board lamination control method in Embodiment 1 also applies to a multilayer rigid-flex circuit board lamination control system, which includes:

[0206] The pressure plate partitioning template 11 is used to perform hierarchical clustering and partitioning of pressure plate based on the pressure plate pressing log to obtain the pressure plate partitioning.

[0207] The platen temperature partitioning module 12 is used to perform platen temperature hierarchical clustering partitioning based on the platen pressing log to obtain platen temperature partitioning.

[0208] The pressure plate target partitioning module 13 is used to take the intersection of the pressure plate pressure partitioning and the pressure plate temperature partitioning to obtain the pressure plate target partitioning;

[0209] The pressure and temperature timing information module 14 is used to receive pressure timing information and temperature timing information, and perform frequent temperature and pressure statistics according to the target partition of the pressure plate to obtain the target partition pressure timing information and the target partition temperature timing information.

[0210] Temperature field timing information module 15 is used to simulate the temperature field of the multilayer rigid-flex circuit board geometry based on the target partition pressure timing information and the target partition temperature timing information to obtain temperature field timing information.

[0211] The pressing control module 16 is used to control the pressure plate to perform pressing control according to the pressure timing information and the temperature timing information when the temperature field timing information is consistent with the standard temperature field timing information.

[0212] Furthermore, the pressure plate partition template 11 is also used for:

[0213] Obtain the initial and target geometry of the multilayer rigid-flex circuit board geometry;

[0214] Retrieve the target position plate pressure plate pressing log that satisfies the pressure plate pressing log of the starting geometry and the target geometry;

[0215] From the pressure plate pressing log, extract the first pressure distribution record information up to the Nth pressure distribution record information that has the same temperature timing record information;

[0216] Based on the first pressure distribution record information up to the Nth pressure distribution record information, pressure plate pressure hierarchical clustering and partitioning are performed to obtain pressure plate pressure partitions.

[0217] Furthermore, the pressure plate partition template 11 is also used for:

[0218] Perform neighborhood hierarchical clustering analysis on the first pressure distribution record information to obtain the first pressure distribution partition;

[0219] Until the neighborhood hierarchical clustering analysis is performed on the Nth pressure distribution record information, the Nth pressure distribution partition is obtained;

[0220] The distribution similarity of the first pressure distribution partition up to the Nth pressure distribution partition is evaluated pairwise to obtain several partition similarities.

[0221] Based on the similarity of the several partitions, outlier partitions are deleted from the first pressure distribution partition up to the Nth pressure distribution partition to obtain a concentrated pressure distribution partition. The intersection of the concentrated pressure distribution partitions is taken as the pressure plate partition.

[0222] Furthermore, the pressure plate partition template 11 is also used for:

[0223] Connect the centers of the first pressure distribution zones to obtain the first pattern;

[0224] Connect the centers of the second pressure distribution zones to obtain the second pattern;

[0225] Calculate the similarity between the first pattern and the second pattern to obtain the first partition similarity, and add it to the plurality of partition similarities.

[0226] Furthermore, the pressure plate temperature partitioning module 12 and the pressure plate target partitioning module 13 can execute the pressure plate pressure partitioning module 11.

[0227] Furthermore, the pressure and temperature timing information module 14 is also used for:

[0228] Constrained by the multilayer rigid-flex circuit board model and the service life of the laminator, the temperature timing information set of the first partition up to the temperature timing information set of the Mth partition, and the pressure timing information set of the first partition up to the pressure timing information set of the Mth partition are collected.

[0229] Simultaneous moment-based value evaluation is performed on the temperature time-series information set of the first partition up to the temperature time-series information set of the Mth partition, and the pressure time-series information set of the first partition up to the pressure time-series information set of the Mth partition, respectively, to obtain the pressure time-series information and temperature time-series information of the target partition.

[0230] Furthermore, the temperature field timing information module 15 is also used for:

[0231] Based on the target partition of the pressure plate, a pressure plate mesh topology is constructed. Combined with the multi-layer rigid-flex circuit board mesh, a topological twin is generated based on a graph neural network to build a geometric temperature field fitting model architecture. The input node of the geometric temperature field fitting model architecture is the target partition of the pressure plate. The input data of any partition is the pressure time series information and the temperature time series information. The output node is the mesh temperature field of the multi-layer rigid-flex circuit board mesh. The multi-layer rigid-flex circuit board mesh will change from the initial mesh to the target mesh according to a preset shrinkage speed.

[0232] Using the multilayer rigid-flex circuit board model and the laminator model as constraints, pressure time-series record information, temperature time-series record information and temperature field time-series record information of the first multilayer rigid-flex circuit board geometry are collected when the service time is less than or equal to the service time threshold. The geometry temperature field fitting model architecture is trained to obtain the temperature field fitting base model.

[0233] Using the multilayer rigid-flex circuit board model and laminating machine location as constraints, the target partition pressure timing record information, the target partition temperature timing record information, and the second multilayer rigid-flex circuit board geometry temperature field timing record information are collected. The temperature field fitting basis model is trained to obtain the temperature field fitting model. Based on the target partition pressure timing information and the target partition temperature timing information, the temperature field of the multilayer rigid-flex circuit board geometry is simulated to obtain the temperature field timing information.

[0234] Furthermore, the pressing control module 16 is also used for:

[0235] When the temperature field time series information is inconsistent with the standard temperature field time series information, the pressure time series information and the temperature time series information are updated, and the updated pressure time series information and the updated temperature time series information are obtained to perform cyclic analysis. At the same time, the pressure time series information and the temperature time series information are stored as the temperature field intervention medium.

[0236] When the number of temperature field intervention media is greater than or equal to the threshold number of intervention media, the deviation parameter between the temperature field intervention media and the standard temperature field intervention media is used as the fitness function to obtain the first temperature field intervention media, the second temperature field intervention media, and the third temperature field intervention media with fitness ranging from small to large.

[0237] With the goal of simultaneously reducing the three deviation parameters of the first temperature field intervention medium, the second temperature field intervention medium, and the third temperature field intervention medium, other temperature field intervention media are updated to obtain expanded pressure time series information and expanded temperature time series information for cyclic analysis.

[0238] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0239] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0240] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0241] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0242] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0243] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0244] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for controlling the pressing of a multilayer rigid-flex circuit board, characterized by, The method comprises the following steps: According to the pressure plate pressing log, the pressure plate pressure hierarchical clustering partition is carried out to obtain the pressure plate pressure partition; According to the pressure plate pressing log, the pressure plate temperature hierarchical clustering partition is carried out to obtain the pressure plate temperature partition; The intersection of the pressure plate pressure partition and the pressure plate temperature partition is obtained to obtain the pressure plate target partition; The pressure time sequence information and the temperature time sequence information are received, and the temperature and pressure frequent statistics are carried out according to the pressure plate target partition to obtain the target partition pressure time sequence information and the target partition temperature time sequence information; Based on the target partition pressure time sequence information and the target partition temperature time sequence information, the temperature field simulation is carried out on the multi-layer rigid-flex circuit board geometry to obtain the temperature field time sequence information; When the temperature field time sequence information is consistent with the standard temperature field time sequence information, the pressure plate is controlled to perform the pressing control according to the pressure time sequence information and the temperature time sequence information.

2. The method of claim 1, wherein, According to the pressure plate pressing log, the pressure plate pressure hierarchical clustering partition is carried out to obtain the pressure plate pressure partition, which comprises the following steps: The starting geometry and the target geometry of the multi-layer rigid-flex circuit board geometry are obtained; The pressure plate pressing log of the target site number pressure plate satisfying the starting geometry and the target geometry is searched; From the pressure plate pressing log, the first pressure distribution record information to the Nth pressure distribution record information with the same temperature time sequence record information is extracted; Based on the first pressure distribution record information to the Nth pressure distribution record information, the pressure plate pressure hierarchical clustering partition is carried out to obtain the pressure plate pressure partition.

3. The method of claim 2, wherein, Based on the first pressure distribution record information to the Nth pressure distribution record information, the pressure plate pressure hierarchical clustering partition is carried out to obtain the pressure plate pressure partition, which comprises the following steps: The first pressure distribution record information is subjected to neighborhood hierarchical clustering analysis to obtain the first pressure distribution partition; Until the Nth pressure distribution record information is subjected to neighborhood hierarchical clustering analysis to obtain the Nth pressure distribution partition; The distribution similarity of the first pressure distribution partition to the Nth pressure distribution partition is evaluated in pairs to obtain a plurality of partition similarities; According to the plurality of partition similarities, the first pressure distribution partition to the Nth pressure distribution partition is subjected to outlier distribution partition deletion to obtain a concentrated pressure distribution partition, and the intersection of the concentrated pressure distribution partition is taken as the pressure plate pressure partition.

4. The method of claim 3, wherein, The distribution similarity of the first pressure distribution partition to the Nth pressure distribution partition is evaluated in pairs to obtain a plurality of partition similarities, which comprises the following steps: The first pressure distribution partition is subjected to partition center connection to obtain a first pattern; The second pressure distribution partition is subjected to partition center connection to obtain a second pattern; The similarity of the first pattern and the second pattern is calculated to obtain a first partition similarity, which is added to the plurality of partition similarities.

5. The method of claim 1, wherein, The pressure time sequence information and the temperature time sequence information are received, and the temperature and pressure frequent statistics are carried out according to the pressure plate target partition to obtain the target partition pressure time sequence information and the target partition temperature time sequence information, which comprises the following steps: Collecting first partition temperature time series information set to Mth partition temperature time series information set and first partition pressure time series information set to Mth partition pressure time series information set satisfying the pressure time series information and the temperature time series information as constraints; Respectively, the first partition temperature time series information set to the Mth partition temperature time series information set, the first partition pressure time series information set to the Mth partition pressure time series information set are evaluated in the simultaneous time set value, and the target partition pressure time series information and the target partition temperature time series information are obtained.

6. The method of claim 1, wherein, Based on the target partition pressure time series information and the target partition temperature time series information, temperature field simulation is carried out on the multi-layer rigid-flex circuit board geometry, and the temperature field time series information is obtained, including: Based on the target partition pressure, the plate grid topology is constructed, combined with the multi-layer rigid-flex circuit board grid body, the topology twin is carried out based on the graph neural network, and the geometry temperature field fitting model architecture is built, wherein the input node of the geometry temperature field fitting model architecture is the target partition of the pressing plate, the input data of any one partition is the pressure time series information and the temperature time series information, and the output node is the grid temperature field of the multi-layer rigid-flex circuit board grid body. The multi-layer rigid-flex circuit board grid body will change from the starting grid body to the target grid body according to the preset reduction speed; With the multi-layer rigid-flex circuit board model and the pressing machine model as constraints, the pressure time series record information, the temperature time series record information and the first multi-layer rigid-flex circuit board geometry temperature field time series record information with service life less than or equal to the service life threshold are collected, the geometry temperature field fitting model architecture is trained, and the temperature field fitting base model is obtained. With the multi-layer rigid-flex circuit board model and the pressing machine model as constraints, the target partition pressure time series record information, the target partition temperature time series record information and the second multi-layer rigid-flex circuit board geometry temperature field time series record information are collected, the temperature field fitting base model is trained, and the temperature field fitting model is obtained. Based on the target partition pressure time series information and the target partition temperature time series information, the temperature field simulation is carried out on the multi-layer rigid-flex circuit board geometry, and the temperature field time series information is obtained.

7. The method of claim 1, wherein, Also includes: When the temperature field time series information is inconsistent with the standard temperature field time series information, the pressure time series information and the temperature time series information are updated to obtain updated pressure time series information and updated temperature time series information for cyclic analysis, and the pressure time series information and the temperature time series information are stored as temperature field intervention medium; When the number of temperature field intervention media is greater than or equal to the number of intervention media threshold, the deviation parameters of the temperature field intervention medium and the standard temperature field intervention medium are used as the fitness function to obtain the first temperature field intervention medium, the second temperature field intervention medium and the third temperature field intervention medium with fitness from small to large; The other temperature field intervention media are updated by simultaneously reducing the three deviation parameters of the first temperature field intervention medium, the second temperature field intervention medium and the third temperature field intervention medium to the target, and the expanded pressure time series information and the expanded temperature time series information are obtained for cyclic analysis.

8. A multi-layer rigid-flex circuit board press control system, characterized by, The system is used for executing the method in any one of claims 1-7, and the system comprises: a press plate pressure partition template, configured to perform press plate pressure hierarchical clustering partition according to a press plate pressing log, and obtain press plate pressure partitions; a press plate temperature partition module, configured to perform press plate temperature hierarchical clustering partition according to the press plate pressing log, and obtain press plate temperature partitions; a press plate target partition module, configured to perform intersection on the press plate pressure partitions and the press plate temperature partitions, and obtain press plate target partitions; a pressure and temperature time sequence information module, configured to receive pressure time sequence information and temperature time sequence information, perform temperature and pressure frequency statistics according to the press plate target partitions, and obtain target partition pressure time sequence information and target partition temperature time sequence information; a temperature field time sequence information module, configured to perform temperature field simulation on a multilayer rigid-flex circuit board geometry based on the target partition pressure time sequence information and the target partition temperature time sequence information, and obtain temperature field time sequence information; a pressing control module, configured to control the press plate to perform pressing control when the temperature field time sequence information is consistent with standard temperature field time sequence information, according to the pressure time sequence information and the temperature time sequence information.

Images