A master-centric multi-carrier dynamic binding management method and system

Abstract

The application provides a kind of main material centralization multi-carrier dynamic binding management method and system, belong to semiconductor manufacturing technical field, this method includes: obtaining main material, first auxiliary material and second auxiliary material and each identification information;Based on the judgment result of the current surface atlas of main material and the preset standard process surface, select bonding or flip;If select bonding, in response to first bonding signal, bond main material and first auxiliary material to obtain first combination material, bind the identification information of both to generate first combination material identification;In response to second bonding signal, bond first combination material and second auxiliary material to obtain second combination material, bind the identification information of both to obtain second combination material identification;In response to first debonding signal, the first auxiliary material in second combination material is debonded, and the main material after debonding is flipped to obtain third combination material;In response to second debonding signal, the second auxiliary material in third combination material is debonded to obtain target main material.

Description

Technical Field

[0001] This application relates to the field of semiconductor manufacturing technology, and in particular to a method and system for dynamic binding management of multiple carriers with centralized main material. Background Technology

[0002] In precision manufacturing fields such as advanced semiconductor packaging and chip stacking, as device integration increases and dimensions shrink, the processing precision of main materials must be guaranteed. However, existing processes suffer from several drawbacks, including: a lack of unified traceability standards, leading to the risk of mixed materials; reliance on manual or simple inspection for identifying the front and back of main materials, resulting in insufficient alignment accuracy and low bonding yield; and issues such as material posture deviation and broken identification links during multi-carrier step-by-step bonding and debonding, impacting production efficiency. Furthermore, ultra-thin main materials are prone to bending and breakage during processing, requiring temporary support from auxiliary materials, but current technologies cannot achieve dynamic adaptation and binding between auxiliary and main materials, or precise debonding. In addition, the lack of a full-process identification binding mechanism makes process traceability difficult and fails to meet the quality control requirements of precision manufacturing.

[0003] Therefore, there is an urgent need for a multi-carrier dynamic binding management method and system with the main material as the core, in order to solve the technical problem of identification and traceability in the collaborative processing of multiple auxiliary materials. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a method and system for centralized multi-carrier dynamic binding management of main materials.

[0005] A first aspect of this application provides a method for centralized multi-carrier dynamic binding management of main materials, including: Obtain the main material, the first auxiliary material, and the second auxiliary material, and identify the identification information of the main material, the first auxiliary material, and the second auxiliary material; Based on the judgment result of the current surface pattern of the main material and the preset standard process surface, the process can be selected to enter the bonding process or control the flipping mechanism to perform the flipping. If the bonding process is selected, in response to the first bonding signal of the device, the main material and the first auxiliary material are physically bonded to obtain the first combined material; the identification information of the main material is bound with the identification information of the first auxiliary material to obtain the identification of the first combined material. In response to the second bonding signal of the device, the first composite material and the second auxiliary material are physically bonded to obtain the second composite material, and the identifier of the first composite material is combined with the identifier information of the second auxiliary material to obtain the identifier of the second composite material; In response to the first debonding signal of the device, the first auxiliary material in the second composite material is debonded, and the debonded main material is flipped to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third composite material is obtained. In response to a second debonding signal from the device, the second auxiliary material in the third composite material is debonded to obtain the target main material. A second aspect of this application provides a main material-centric multi-carrier dynamic bonding management system, including: The material and identification information acquisition module is used to acquire the main material, the first auxiliary material, and the second auxiliary material, and to identify the identification information of the main material, the first auxiliary material, and the second auxiliary material; The process surface judgment and process selection module is used to select to enter the bonding process or control the flipping mechanism to perform flipping based on the judgment result of the current surface pattern of the main material and the preset standard process surface. The first bonding and identification binding module is used to, in response to the first bonding signal of the device, physically bond the main material and the first auxiliary material to obtain the first combined material if the bonding process is selected to enter; and bind the identification information of the main material and the identification information of the first auxiliary material to obtain the identification of the first combined material. The second bonding and identification combination module is used to physically bond the first composite material and the second auxiliary material in response to the second bonding signal of the device to obtain the second composite material, and combine the identification information of the first composite material with the identification information of the second auxiliary material to obtain the identification information of the second composite material. The first debonding and main material flipping module is used to debond the first auxiliary material in the second combined material in response to the first debonding signal of the device, and flip the debonded main material to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third combined material is obtained. The second debonding and target material acquisition module is used to debond the second auxiliary material in the third combined material in response to the second debonding signal of the device to obtain the target main material. A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the above-described main material-centralized multi-carrier dynamic binding management method.

[0006] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described material-centric multi-carrier dynamic binding management method. The beneficial effects of the material-centric multi-carrier dynamic binding management method and system provided by this application are as follows: By identifying the identification information of the main material and each auxiliary material, this application establishes a full-process identification binding system with the main material as the core, achieving accurate traceability of material associations during multi-carrier processing, effectively avoiding problems such as material mixing and identification breakage, and further ensuring the traceability and controllability of the production process.

[0007] Based on the comparison between the current surface pattern of the main material and the standard process surface, the flipping or bonding process is triggered, improving the attitude calibration accuracy of the main material. The step-by-step physical bonding and identification binding of the main material with the first and second auxiliary materials ensures the orderly processing of multiple carriers and further adapts to complex process requirements. This application achieves dynamic control of the entire process from bonding to debonding of multiple carriers, improving the processing efficiency and operational flexibility of precision manufacturing, strengthening quality control capabilities, and providing reliable technical support for the production of highly integrated, high-precision devices. Attached Figure Description

[0008] Figure 1 A flowchart illustrating a centralized multi-carrier dynamic binding management method for main materials provided in an embodiment of this application; Figure 2 A structural block diagram of a main material-centralized multi-carrier dynamic binding management system provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0009] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0010] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1-3 The following is an explanation using specific examples.

[0011] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a centralized multi-carrier dynamic binding management method for main materials provided in an embodiment of this application. The method includes: S101: Obtain the main material, the first auxiliary material, and the second auxiliary material, and identify the identification information of the main material, the first auxiliary material, and the second auxiliary material.

[0012] In this embodiment, the primary task in the initial stage of the multi-carrier combination process is to accurately acquire and verify the identities of the main and auxiliary materials. The main material is the core carrier of the entire process; its acquisition requires coordination with the production plan and material scheduling system, retrieving it from designated warehouses or turnover stations. During the retrieval process, the batch information and pre-processing status (whether it has undergone cleaning, pretreatment, etc.) of the main material must be confirmed through the warehouse management module. The first and second auxiliary materials are matched according to the process requirements of the main material to avoid incompatibility between the auxiliary material's model and specifications and the main material.

[0013] The core objective of identification information is to assign a unique digital identity to each material. The identification information for the main material, the first auxiliary material, and the second auxiliary material is carried out in the form of serial numbers or RFID tags. The main material identification includes a unique material code, material type, dimensional parameters, production batch, and quality inspection results; the identification information for the first and second auxiliary materials includes the auxiliary material code, expiration date, compatible main material model, and supplier information.

[0014] The identification process uses visual recognition devices (cameras) or RFID tag readers to read the marking information on the material surface. The read marking data is transmitted to the production execution system in real time. The production execution system analyzes and compares the data to confirm the legality of the material's identity, its match with the production task, and whether there are any abnormalities such as expiration or non-compliance. If problems such as unreadable tags or inconsistent data comparison occur during the identification process, an alarm mechanism is triggered to notify staff for manual intervention.

[0015] S102: Based on the judgment result of the current surface pattern of the main material and the preset standard process surface, select to enter the bonding process or control the flipping mechanism to perform flipping.

[0016] In this embodiment, the acquisition of the current surface map is based on a visual inspection system, which consists of an industrial camera, a light source module, an image acquisition card, and image processing software. After the main material is transferred to the designated inspection station, the light source module provides a uniform, glare-free lighting environment, while the industrial camera performs high-definition imaging of the main material surface. The acquired image data is transmitted to the processing software via the image acquisition card to generate a current surface map including information such as surface texture, feature markers, and geometric contours. The preset standard process surface is a template based on product design requirements and process parameter standards. This standard process surface needs to be pre-entered into the system's database, which includes key feature parameters of the specific surfaces of the main material used for bonding, such as feature point coordinates, texture density, and marker shape, as a basis for comparison and judgment.

[0017] Specifically, the current bonding surface map is compared with the standard process surface map for feature extraction and similarity calculation. The key feature parameters of the two maps are compared to determine if they are consistent. Simultaneously, any attitude deviations caused by material placement errors or slight shifts during transport are identified. If the comparison results show that the key features of the current bonding surface map match the standard process surface map, and the attitude deviation is within acceptable limits, it indicates that the current bonding surface of the main material is the bonding surface that meets the process requirements. A bonding permission command is then generated, triggering the initiation of subsequent bonding processes.

[0018] If the comparison results show that the current facet pattern does not match the standard process facet, or the attitude deviation exceeds the preset allowable threshold, it indicates that the current facet of the main material is not the target process facet, or that there is an attitude abnormality, thus triggering the working command of the flipping mechanism. The allowable threshold value is determined based on the material composition, dimensional accuracy, bonding process requirements, and historical production data of the main material.

[0019] S103: If the bonding process is selected, in response to the first bonding signal of the device, the main material and the first auxiliary material are physically bonded to obtain the first combined material; the identification information of the main material is bound with the identification information of the first auxiliary material to obtain the identification of the first combined material.

[0020] In this embodiment, once the main material is confirmed to meet the requirements of the standard process surface by the surface determination, it enters the bonding process ready state. Specifically, in response to the first bonding signal of the equipment, the production execution system client is triggered to send a first verification request for the legality of the operation to the production execution system server. When the production execution system server receives the first verification request, it performs verification based on a preset multi-dimensional verification rule set to obtain the verification result. When the verification result is that all rules in the multi-dimensional verification rule set meet the preset bonding conditions, the main material and the first auxiliary material are physically bonded to obtain the first combined material. When any rule in the multi-dimensional verification rule set does not meet the bonding conditions, the physical bonding of the main material and the first auxiliary material is prevented.

[0021] Simultaneously with the completion of physical bonding, the binding process of identification information proceeds concurrently. Specifically, using the identification information of the primary material as an index, the identification information of the first auxiliary material is bound to the primary material's identification through an association algorithm, generating a unique first combined material identification. This first combined material identification is not a simple aggregation of information, but a comprehensive data carrier encompassing multiple dimensions of information, including basic information of the primary and first auxiliary materials (material, specifications, batch), bonding process parameters (bonding pressure, temperature, time), bonding process data (alignment accuracy, monitoring data), and quality inspection results. After binding, the first combined material identification is uploaded to the database of the production execution system.

[0022] The operations at the temporary bonding site include: placing the silicon interposer main material wafer cassette and the first auxiliary material (temporary bonding glass) wafer cassette on the equipment; the equipment identifies the wafer cassette number and uploads the wafer cassette number to the equipment automation protocol; the equipment automation protocol performs operation legality verification, which includes verification of the current status of the material (must be in a waiting state), process path verification (must be at the temporary bonding site), batch consistency verification, historical bonding relationship verification (unbonded state), and physical parameter consistency verification.

[0023] For example, when all rules meet the preset conditions, the equipment is allowed to perform the operation; when any rule is not met, a rejection signal is sent and state transition is prevented; the equipment opens the wafer cassette, takes out the wafer, and places the wafer in the optical character recognition unit to read the wafer serial number; the equipment sends the serial number information to the equipment automation protocol, and queries the manufacturing execution system information through the equipment automation protocol; the equipment verifies the data through the manufacturing execution system, including comparing the temporary bonding glass serial number / wafer serial number with the manufacturing execution system data to confirm that it is indeed this batch of materials. After the data is correct, the equipment is instructed to perform bonding; the equipment performs the wafer bonding action; after the equipment completes the physical bonding, it sends a completion signal to the equipment automation protocol; the equipment automation protocol transmits the data to the manufacturing execution system through the network application programming interface; after receiving the data, the manufacturing execution system binds the serial number and updates the database; and generates new accompanying document information synchronously.

[0024] S104: In response to the second bonding signal of the device, the first composite material and the second auxiliary material are physically bonded to obtain the second composite material, and the identifier of the first composite material is combined with the identifier information of the second auxiliary material to obtain the identifier of the second composite material.

[0025] In this embodiment, in response to the second bonding signal from the device, a first bonding process file associated with the first composite material identifier is retrieved. Based on the first bonding process file, potential weak areas are predicted when the first composite material identifier and the second auxiliary material are bonded, and a second non-uniform bonding parameter matrix for the second auxiliary material is generated. The bonding head is controlled based on the second non-uniform bonding parameter matrix to physically bond the first composite material and the second auxiliary material, resulting in the second composite material. Simultaneously, the identifier of the first composite material and the identifier information of the second auxiliary material are combined to obtain the identifier of the second composite material.

[0026] S105: In response to the first debonding signal of the device, the first auxiliary material in the second composite material is debonded, and the main material after debonding is flipped to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third composite material is obtained.

[0027] In this embodiment, in response to the first debonding signal from the device, the Manufacturing Execution System (MES) client is triggered to send a second verification request to the Manufacturing Execution System (MES) server to verify the legality of the debonding operation. The MES server performs verification based on a preset debonding verification rule set, which includes: material current status verification, process path verification, historical binding relationship verification, batch consistency verification, and physical parameter consistency verification.

[0028] Specifically, when all rules in the verification rule set meet the preset debonding conditions, the first auxiliary material in the assembly is debonded, and the debonded assembly is flipped to obtain the third composite material.

[0029] The operations at the debonding station include: placing the wafer cassette to be debonded onto the equipment; the equipment identifying the wafer cassette number and uploading the cassette number information to the equipment automation protocol; the equipment automation protocol performing job legality verification, including: material current status verification (must be in a pending operation state), process path verification (must be at the debonding station), historical binding relationship verification (bound state), and physical parameter consistency verification. Specifically, when all the above conditions are met, the equipment is allowed to execute the job; when any rule is not met, a rejection signal is sent and state transition is prevented; the equipment opens the wafer cassette, removes the wafer, and places the wafer in the optical character recognition unit to read the wafer serial number; the equipment sends the serial number information to the equipment automation protocol, which queries the manufacturing execution system information; and the manufacturing execution system data is verified, including comparing the temporary bonding glass serial number / wafer serial number with the manufacturing execution system data to confirm that it is indeed from this batch of materials. After confirming the data is correct, the machine is instructed to debond; the machine performs the debonding action; after completing the debonding, the machine sends a signal to the machine automation protocol; the machine automation protocol transmits the data to the manufacturing execution system through the interface; after receiving the data, the manufacturing execution system updates the status and restores the master serial number.

[0030] S106: In response to the second debonding signal of the device, the second auxiliary material in the third composite material is debonded to obtain the target main material.

[0031] In this embodiment, the triggering of the second debonding signal is a crucial node connecting the first debonding and the acquisition of the target main material. Its core is that after the third composite material completes the peeling and surface calibration of the first auxiliary material, the separation process of the second auxiliary material is initiated via a command. The generation of this command undergoes multiple precondition checks. Specifically, the production execution system first confirms that the identification information of the third composite material matches the debonding task sheet, the surface condition of the main material after the first debonding meets the benchmark requirements, the equipment status of the debonding station (debonding head calibration accuracy, energy output stability) meets the standards, and the environmental cleanliness meets the requirement of no particulate contamination. After all conditions are met, the system sends the second debonding signal to the equipment, simultaneously carrying specific debonding parameters for the second auxiliary material (separation energy, temperature threshold, force direction, etc.).

[0032] The debonding process of the second auxiliary material is based on the structural characteristics and bonding method of the third composite material, employing precise and controllable separation technology to achieve efficient peeling. Specifically, firstly, preset positioning parameters are retrieved based on the identification information of the third composite material, and the position of the composite material is calibrated through a visual inspection system. Subsequently, the debonding equipment initiates the corresponding separation mode based on the process parameters carried by the second debonding signal. For example, if the auxiliary material is thermo-bonded, the adhesive layer of the bonding surface will be softened through precise temperature control; if the auxiliary material is ultrasonically bonded, the physical connection of the bonding surface will be destroyed through vibration energy of a specific frequency; if it is a mechanical snap-fit ​​connection, non-destructive separation will be achieved through a micro-force drive mechanism. During the separation process, the equipment's force feedback sensor and visual inspection system continuously work, collecting data such as the magnitude of the separation force, the peeling speed of the auxiliary material, and the surface state of the main material, adjusting the debonding energy and the direction of the force.

[0033] Once the second auxiliary material is completely peeled off, the target main material is obtained. At this point, a multi-dimensional quality verification process is required to confirm whether it meets the process requirements. First, a visual inspection system performs a comprehensive imaging of the surface of the target main material to check for scratches, dents, particle residue, functional layer detachment, and other damage. Simultaneously, it compares the surface images before and after peeling to confirm that the geometric shape and dimensional accuracy of the main material have not changed abnormally. Second, performance testing equipment performs sampling tests on the core performance parameters of the main material (hardness, flatness, electrical / thermal conductivity, etc.) to verify that the debonding process has not affected its core functions. If the test reveals surface damage or substandard performance of the main material, it is marked as a non-conforming product and transferred to the rework or scrap process; if the test results meet the preset standards, a conformity report for the target main material is generated.

[0034] In another embodiment, the FOI segment is the wafer-level packaging front-end process. First, the silicon interposer main material is batched and the metal bump fabrication process is completed. Simultaneously, two temporary bonding glass auxiliary materials, the first auxiliary material and the second auxiliary material, complete the pre-processing of 220 photoresist coating and aluminum sputtering physical vapor deposition. Then, the first auxiliary material and the silicon interposer main material are temporarily bonded to form a silicon interposer main material and the first auxiliary material combination. The system binds the serial numbers of the two materials and the operator prints the corresponding accompanying documents. This combination continues to complete the back via exposure and controlled collapse chip connection process. Then, it is temporarily bonded to the second auxiliary material to form a silicon interposer main material + first auxiliary material + second auxiliary material combination. The system updates the serial number binding information and reprints the accompanying documents. Finally, the silicon interposer main material + first auxiliary material + second auxiliary material combination is debonded and the first auxiliary material is removed. The resulting silicon interposer main material + second auxiliary material combination is then put into storage.

[0035] The FOP (Fold-on-Package) stage is the back-end process of wafer-level packaging. First, the silicon interposer main material + second auxiliary material combination that has been put into the warehouse is batched with the bare chip dies prepared by the back metallization process. The batched materials sequentially complete the entire chip interconnect process of flip chip mounting, underfill, and chip mounting. Then, the silicon interposer main material + second auxiliary material combination is debonded and the second auxiliary material is removed. After removing all auxiliary materials, the remaining silicon interposer main material continues to be processed.

[0036] As can be seen from the above, this application establishes a full-process identification and binding system centered on the main material by identifying the identification information of the main material and each auxiliary material. This achieves accurate traceability of material associations during multi-carrier processing, effectively avoiding problems such as material mixing and identification breakage, and further ensuring the traceability and controllability of the production process. Furthermore, by comparing the current surface pattern of the main material with the standard process surface, a flipping or bonding process is triggered, improving the attitude calibration accuracy of the main material. The step-by-step execution of physical bonding and identification combination binding between the main material and the first and second auxiliary materials ensures the orderly processing of multiple carriers and further adapts to complex process requirements. Overall, this application achieves dynamic control of the entire process from bonding to debonding of multiple carriers, improving the processing efficiency and operational flexibility of precision manufacturing, strengthening quality control capabilities, and providing reliable technical support for the production of highly integrated, high-precision devices.

[0037] In one embodiment of this application, based on the judgment result of the current surface pattern of the main material and the preset standard process surface, the method selects to enter the bonding process or controls the flipping mechanism to perform flipping, including: The robotic arm in the equipment picks up the main material from the material box and transfers it to the flipping station; The first surface pattern is obtained by acquiring the current surface pattern of the main material through the optical sensor on the flipping station; Based on the identification information of the main material, determine whether the first surface pattern is a standard process surface, and obtain the judgment result; If the determination result is that the first pattern is a preset standard process surface, then it is bonded to the first auxiliary material; If the judgment result is that the first side of the pattern is not a standard process surface, then control the flipping mechanism to perform flipping and alignment actions to obtain the second side of the pattern.

[0038] In this embodiment, after receiving a gripping command, the robotic arm picks up the main material from the designated material box. The picking process confirms the material model and batch match the current production task based on the main material's identification information, and ensures stable, collision-free, and offset picking through a force control or vision inspection system. Subsequently, the robotic arm moves the main material to the flipping station according to a preset path, placing the main material within the detection area of ​​the optical sensor imaging. The preset path is pre-planned based on the main material size, equipment layout, station coordinates, and process safety distances.

[0039] At the flipping station, an optical sensor images the surface of the main material, acquiring and generating a first surface map. This first surface map includes not only visual information such as surface texture, feature markers, and structural contours, but also information such as flatness and feature point coordinates. Based on the identified main material identification information, the corresponding standard process surface is retrieved as a benchmark. The first surface map is then compared with the standard process surface in terms of feature matching, orientation, and consistency, ultimately determining whether it is a standard process surface or not.

[0040] Specifically, if the judgment result indicates that the first surface pattern is a standard process surface, it means that the current posture and orientation of the main material meet the process requirements. In this case, the main material is transferred to the bonding station to prepare for bonding with the first auxiliary material. If the judgment result indicates that the current surface is not a standard process surface, bonding is not initiated, and a control command is sent to the flipping mechanism to drive it to perform flipping and alignment actions on the main material to obtain the second surface pattern.

[0041] As can be seen from the above, this embodiment, by using a gripping robotic arm to pick up and transfer materials, and by using optical sensors to collect surface maps and compare them with standard process surfaces, can accurately identify the front and back sides and orientation of the main material before bonding, and select bonding or flipping alignment. This effectively avoids bonding failure and product damage caused by incorrect orientation or positional offset of the main material, further improving the automation level and production stability of the process, while reducing manual intervention and improving production efficiency and product yield.

[0042] In one embodiment of this application, if the determination result is that the first surface pattern is not a preset standard process surface, the flipping mechanism is controlled to perform flipping and alignment actions to obtain the second surface pattern, including: Obtain the deviation angle and / or deviation displacement between the first surface pattern and the standard process surface, and generate the first attitude deviation matrix; The first attitude deviation matrix is ​​input to the motion controller of the flipping mechanism, which drives the flipping mechanism to perform dynamic pose compensation of the main material through multi-axis linkage during the flipping process, so as to obtain the flipped main material. The image data of the flipped main material is obtained to obtain the second-side pattern.

[0043] In this embodiment, by performing image feature matching and coordinate calculation on the first surface map and the standard process surface, the deviation angle and deviation displacement of the main material at the current workstation are extracted, and these discrete deviation information are integrated into a first attitude deviation matrix. This first attitude deviation matrix describes the attitude error of the main material in space in digital form, and the attitude error includes multi-dimensional deviation information such as rotation, translation, and tilt.

[0044] After obtaining the attitude deviation matrix, it is sent to the motion controller of the flipping mechanism. The motion controller plans the motion trajectory according to the deviation matrix and drives the flipping mechanism to perform real-time and dynamic posture compensation of the main material through multi-axis linkage while performing the flipping action, so that the main material reaches the standard posture required by the process at the same time as the flipping is completed.

[0045] After flipping and dynamic compensation are completed, image data of the main material surface is acquired again using an optical sensor to generate a second surface map. This second surface map is used to reconfirm whether the surface orientation and attitude of the main material meet the standard process surface requirements, providing a final judgment basis for whether to proceed to the bonding process, thus forming an automated closed loop from detection to calculation to compensation to verification.

[0046] As can be seen from the above, this embodiment generates an attitude deviation matrix by calculating the deviation between the first surface pattern and the standard process surface, and then realizes dynamic pose compensation during the flipping process based on multi-axis linkage. This can improve the accuracy of the correction angle and displacement deviation while completing the flipping of the main material, effectively improve the accuracy of the flipping and alignment of the main material, and further avoid the problems of large positioning error and low efficiency of repeated alignment in the existing flipping method. This effectively ensures that the attitude of the main material meets the requirements of the subsequent bonding process.

[0047] In one embodiment of this application, a multi-carrier dynamic binding management method with centralized main material further includes: Retrieve the reference map corresponding to the preset standard process surface, and perform difference calculation between the second surface map and the reference map to generate a difference image; Based on differential imaging, particulate contamination or mechanical damage to the host material is identified, and based on this particulate contamination or mechanical damage, bonding pressure parameters or ultrasonic energy parameters for the bonding process are determined, including: Pixel-level segmentation is performed on the differential image to extract the location coordinate set of particulate contamination or mechanical damage in the main material; Map the set of position coordinates to the coordinate system of the bonding head's motion trajectory to generate a defect spatial distribution map; Based on the defect spatial distribution map, the required compensation bonding energy for each bonding site is calculated, and a non-uniform bonding parameter matrix is ​​generated. The non-uniform bonding parameter matrix includes bonding pressure parameters or ultrasonic energy parameters corresponding to the location coordinate set.

[0048] In this embodiment, a reference image of the standard process surface corresponding to the main material is first retrieved. The reference image and the acquired second-surface image are then subjected to pixel-by-pixel differential calculation to generate a differential image that highlights the differences. Through this differential image, subtle defects such as particulate contamination, scratches, dents, and localized deformations, which were originally obscured by surface texture and normal structure, are enhanced, thereby achieving high-sensitivity identification of surface defects. Based on this, pixel-level segmentation processing is performed on the differential image to locate the coordinates, size, and shape of each defect, forming a set of defect location coordinates.

[0049] The location coordinates are mapped from the image coordinate system to the motion trajectory coordinate system of the bonding device, generating a spatial distribution map representing the defect distribution. Based on this map, the compensation energy required for different bonding positions is calculated point-by-point. Specifically, bonding pressure or ultrasonic energy is appropriately adjusted in defective areas to avoid excessive pressure causing further damage or insufficient energy leading to weak bonding; in intact areas, standard parameters are used to protect the substrate surface while ensuring bonding strength.

[0050] Finally, a non-uniform bonding parameter matrix is ​​generated that corresponds one-to-one with the surface state of the host material. This non-uniform bonding parameter matrix includes bonding pressure parameters or ultrasonic energy parameters corresponding to each coordinate position, enabling the bonding head to perform differentiated bonding actions in different regions.

[0051] As can be seen from the above, this embodiment improves the accuracy of identifying the location of particulate contamination and mechanical damage on the surface of the main material by performing differential calculation between the second surface map and the reference map, segmenting defects at the pixel level, and mapping them to generate a defect spatial distribution map. This improves the accuracy of compensating for bonding energy. The resulting non-uniform bonding parameter matrix effectively avoids weak bonding, over-soldering, or damage to the main material caused by surface defects by using differentiated bonding pressure or ultrasonic energy for different regions during the bonding process. This improves bonding quality, process stability, and product yield, while also achieving intelligent, refined, and adaptive control of bonding parameters.

[0052] In one embodiment of this application, in response to a first bonding signal from the device, a main material and a first auxiliary material are physically bonded to obtain a first composite material, comprising: In response to the device's first bonding signal, the production execution system client is triggered to send the first verification request for job validity to the production execution system server; The production execution system server receives the first verification request and performs verification based on a preset multi-dimensional verification rule set to obtain the verification result; When the verification result shows that all rules in the multi-dimensional verification rule set meet the preset bonding conditions, the main material and the first auxiliary material are physically bonded to obtain the first combined material. If any rule in the multi-dimensional verification rule set fails to meet the bonding conditions, then physical bonding between the main material and the first auxiliary material is prevented.

[0053] In this embodiment, when the device sends the first bonding signal, the bonding action is not initiated. Instead, the production execution system client is first triggered to send a first verification request for the legality of the job to the production execution system server. This first verification request carries information such as the identification information of the main material and the first auxiliary material, the current workstation number, the process step number, and the equipment status.

[0054] Upon receiving a verification request, the Manufacturing Execution System (MES) server performs a comprehensive verification based on a pre-set multi-dimensional verification rule set. This rule set includes material status verification, process path verification, batch consistency verification, historical binding relationship verification, and physical parameter consistency verification. Only when all verification rules meet the pre-set bonding conditions is the physical bonding of the primary material and the first auxiliary material initiated, forming the first combined material. The pre-set bonding conditions are that all four verifications—material status verification, process path verification, batch consistency verification, historical binding relationship verification, and physical parameter consistency verification—must pass.

[0055] If any item in the verification rule set fails to meet the bonding conditions, such as status not meeting the standard, material error, skipped process steps, or insufficient permissions, the bonding request is determined to be an illegal operation, and a blocking command is returned to the equipment to forcibly prohibit physical bonding execution. At the same time, the abnormal information is recorded and an alarm is uploaded.

[0056] The verification process is based on a pre-defined multi-dimensional set of verification rules. Specifically, in the material status verification, the identification information of the main material and the first auxiliary material is read to check whether they are in a qualified state awaiting bonding, whether they have expired, whether they are locked or scrapped, so that only valid materials can enter the process. In the process path verification, the current bonding process is compared with the product's pre-defined process route to determine whether there are any violations such as skipping processes, reversing processes, or performing bonding without going through the previous process. In the batch consistency verification, the production batch, work order batch, and material version of the main material and the first auxiliary material are checked to prevent the mixing of materials of different batches and specifications. In the historical binding relationship verification, it is checked whether there are any historical binding records of the main material that have not been properly unbound to avoid duplicate binding or incorrect binding that could lead to data chaos. In the physical parameter consistency verification, the thickness, size, material, surface characteristics, and other parameters of the main material and the first auxiliary material are compared to see if they are compatible. The preset process route is configured in advance in the production execution system, which specifies the standard process flow sequence of the product from feeding, posture detection, flipping, bonding with the first auxiliary material, bonding with the second auxiliary material to step-by-step debonding. It is used to determine whether the current bonding process is compliant and whether there are skipped steps or reverse processes.

[0057] As can be seen from the above, this embodiment triggers the operation legality verification between the production execution system client and server after the first bonding signal response, and performs comprehensive verification based on a multi-dimensional rule set including material status, process path, batch consistency, historical binding relationship, and physical parameter consistency. This effectively intercepts abnormal situations such as invalid materials, non-standard processes, batch mismatch, and inconsistent parameters before physical bonding, avoiding material waste and product scrap caused by incorrect bonding. It further improves the standardization, security, and traceability of the process flow, and ensures the stable and reliable operation of the centralized multi-carrier dynamic binding management of main materials. In one embodiment of this application, a first composite material is obtained by physically bonding a main material and a first auxiliary material, comprising: Based on the bonding pressure parameter or ultrasonic energy parameter in the non-uniform bonding parameter matrix, the motion controller is adjusted to apply bonding energy to the region corresponding to the position coordinate set, and physical bonding with the first auxiliary material is performed. By deploying monitoring sensors in the bonding area, images of the bonding interface or bonding strength characteristic signals of the bonded area are collected. The bonding interface image or bonding strength feature signal is compared with the preset standard bonding quality feature and the deviation is calculated to obtain the quality deviation value. If the quality deviation value is greater than the preset upper limit threshold of the process allowable deviation, the parameter compensation amount is calculated in reverse based on the quality deviation value and used as the correction coefficient. The correction factor is applied to the remaining regions in the non-uniform bonding parameter matrix that have not yet been bonded, and the local pressure parameters or local ultrasonic energy parameters of the remaining regions are updated. The bonding head is controlled based on local pressure parameters or local ultrasonic energy parameters to continue bonding the remaining areas and obtain the first composite material identifier.

[0058] In this embodiment, the physical bonding between the main material and the first auxiliary material does not employ a uniform bonding method with fixed parameters. Instead, it utilizes a non-uniform bonding parameter matrix to achieve regionalized, adaptive, and closed-loop corrective intelligent bonding, further ensuring bonding quality and the safety of the main material surface. Based on the previously identified defect location coordinates, corresponding bonding pressure parameters or ultrasonic energy parameters are invoked to drive the motion controller to apply differentiated bonding energy to different regions. Compensation parameters are used in defective regions, while standard parameters are used in intact regions, thereby ensuring bonding strength while avoiding secondary damage to the main material.

[0059] During the bonding process, monitoring sensors deployed in the bonding area collect real-time interface images or bonding strength characteristic signals of the completed bonding areas. This data is compared with preset standard bonding quality characteristics, and an algorithm calculates the quality deviation value to determine whether the current bonding effect meets the process requirements. When the quality deviation value is less than the preset upper limit threshold for process allowable deviation, it indicates that the current bonding parameters are reasonable, and the bonding action in subsequent areas will continue based on the original non-uniform bonding parameter matrix. The preset standard bonding quality characteristics are a set of bonding quality benchmark characteristics determined after integrating the material properties of the main material and the first auxiliary material, the bonding process requirements of the product design, industry precision manufacturing standards, and through process testing and mass production verification. The upper limit threshold for process allowable deviation is set comprehensively based on the material properties of the main material and the first auxiliary material, the bonding process type (pressure / ultrasonic bonding), the bonding quality standards of the product design, and with reference to historical production bonding quality data and process test verification results.

[0060] If the quality deviation value exceeds the upper limit of the process allowable deviation threshold, it indicates that the current local bonding parameters are insufficient or excessive. Based on this deviation value, a reverse calculation is performed to obtain the corresponding parameter compensation amount, which is then used as a correction coefficient. This correction coefficient is applied to the unbonded regions in the non-uniform bonding parameter matrix, dynamically updating the local pressure parameters or local ultrasonic energy parameters of these unbonded regions. Subsequently, the updated local parameters control the bonding head to perform full-area bonding, obtaining the first composite material, and simultaneously generating a first composite material identifier bound to material information, process parameters, and quality data.

[0061] As can be seen from the above, this embodiment achieves regional application of bonding energy based on a non-uniform bonding parameter matrix, and collects interface images and intensity signals in real time during the bonding process, calculates quality deviations, and generates correction coefficients in reverse to dynamically update the pressure or ultrasonic parameters of unbonded areas. This enables the formation of a closed-loop control system with real-time monitoring and online adaptive correction during the bonding process, effectively improving the stability and uniformity of bonding quality. It further avoids poor bonding, over-soldering, or damage to the main material caused by local defects or unreasonable parameters. At the same time, it provides complete and traceable process data for the identification of the first composite material, thereby improving product yield and the level of intelligence in the process flow.

[0062] In one embodiment of this application, in response to a second bonding signal from the device, a first composite material and a second auxiliary material are physically bonded to obtain a second composite material. The identifier of the first composite material is combined with the identifier information of the second auxiliary material to obtain a second composite material identifier, including: In response to the second bonding signal from the device, the first bonding process file associated with the first composite material identifier is retrieved; Based on the first bonding process file, the potential weak areas when the first composite material identifier is bonded to the second auxiliary material are predicted, and a second non-uniform bonding parameter matrix is ​​generated for the second auxiliary material. Based on the second non-uniform bonding parameter matrix, the bonding head is controlled to physically bond the first composite material and the second auxiliary material to obtain the second composite material. The identifier of the first composite material and the identifier information of the second auxiliary material are physically combined to obtain the identifier of the second composite material.

[0063] In this embodiment, when the device sends a second bonding signal, the default process parameters are not used. Instead, the first bonding process file associated with the first combined material is retrieved from the production execution system based on the first combined material identifier. The first bonding process file records the non-uniform bonding parameter matrix of the previous bonding, real-time monitoring data, dynamic correction records, and the historical distribution of quality deviation values.

[0064] After obtaining the first bonding process file, the structural characteristics, interface state, stress distribution, and quality deviation areas that occurred in the previous process of the first composite material are comprehensively evaluated through analysis and modeling of historical data. This allows for the prediction of potential weak areas, such as the location of surface defects, that may occur during the bonding with the second auxiliary material. Based on these predictions, a second non-uniform bonding parameter matrix adapted to this bonding process is generated. The bonding pressure or ultrasonic energy is appropriately optimized for weak areas, while standard parameters are maintained for stable areas, achieving coordinated matching between the two bonding processes.

[0065] Subsequently, based on the second non-uniform bonding parameter matrix, the bonding head is controlled to complete the physical bonding of the first composite material and the second auxiliary material, forming a more structurally complete second composite material. Simultaneously with the physical bonding, the identifier of the first composite material, which already includes information about the main material and the first auxiliary material, is hierarchically combined and bound with the unique identifier of the second auxiliary material to generate a second composite material identifier. This second composite material identifier not only links the identity information of the three materials but also inherits the process files and quality data from both bonding processes, achieving dynamic binding management centered on the main material and traceable across multiple carriers throughout the entire process. This effectively improves the consistency, stability, and traceability of multi-layer bonded products.

[0066] From the above, it can be concluded that this embodiment, by retrieving the first bonding process file, including the non-uniform bonding parameter matrix, real-time monitoring data, dynamic correction records, and historical distribution of quality deviation values, when responding to the second bonding signal, can improve the accuracy of predicting potential weak areas of secondary bonding based on the previous bonding data. This leads to the generation of an adapted second non-uniform bonding parameter matrix, achieving coherent optimization and adaptive control of the front and rear bonding processes. Simultaneously, combining the first combined material identifier and the second auxiliary material identifier to form the second combined material identifier not only improves the bonding strength and product stability of multilayer bonding but also perfects the multi-carrier full-process traceability system centered on the main material, ensuring that the process is inheritable, the quality is predictable, and the data is correlated.

[0067] In one embodiment of this application, a method for generating a second non-uniform bonding parameter matrix of a second auxiliary material includes: Retrieve the historical bond quality dataset associated with the first composite material identifier. The historical bond quality dataset includes at least one actual bond strength distribution map and corresponding process parameter records recorded in the previous bonding process. Based on historical bonding quality datasets, a Gaussian process regression model is constructed with bonding position coordinates as input and bonding strength deviation as output. The current bonding interface morphology features of the first composite material are input into the Gaussian process regression model to predict the probability distribution of strength weaknesses of the first composite material and the second auxiliary material at multiple bonding sites. Based on the probability distribution of weak strength, a weighted compensation is applied to the preset standard bonding parameter matrix to generate a second non-uniform bonding parameter matrix. At bonding sites where the probability distribution of weak strength is greater than a preset threshold, the applied bonding pressure or ultrasonic energy is greater than a preset standard value.

[0068] In this embodiment, the historical bonding quality dataset associated with the first composite material identifier is first retrieved. This dataset includes at least one actual bonding strength distribution map, bonding position coordinates, process parameter execution records, and quality deviation changes collected in the pre-bonding process.

[0069] Secondly, based on historical bonding quality datasets, a Gaussian process regression model is constructed using bonding location coordinates as model input and the corresponding bonding strength deviation value as model output. This Gaussian process regression model can learn the mapping relationship between location, parameters, and bonding strength in historical processes, and make probabilistic predictions of bonding effects at unknown locations. It is particularly suitable for handling scenarios with uncertainties such as minor defects, uneven stress, and strength fluctuations at the bonding interface, and has higher accuracy and generalization ability compared to calculations using fixed formulas.

[0070] The Gaussian process regression model is a single-input, single-output end-to-end prediction model, consisting of three layers: a data layer, a feature layer, and a prediction layer. The training process uses a historical bond quality dataset for model fitting and validation. Specifically, the data layer preprocesses the retrieved historical bond quality data, removing outliers, filling in missing data, normalizing bond position coordinates, and standardizing bond strength deviation values ​​to form a regular training and validation set. The feature layer extracts spatial features of bond position coordinates, establishes a mapping relationship between position and bond strength deviation, and explores potential patterns between position, process parameters, and bond strength in historical processes. The prediction layer constructs a probabilistic prediction model based on the kernel function of Gaussian process regression, iteratively trains the model using the training set, verifies the model's prediction accuracy using the validation set, and continuously optimizes the kernel function parameters until the model's prediction deviation meets the preset accuracy requirements of the process. After training, the model is embedded into the system for predicting the probability distribution of weak bond strength in actual production.

[0071] The parameter settings of the Gaussian process regression model revolve around three main categories: kernel function parameters, prior parameters, and prediction hyperparameters. All parameters are adapted to the actual characteristics of semiconductor bonding processes and the distribution patterns of historical data. Specifically, the kernel function uses a squared exponential kernel function suitable for spatial location prediction. Its length scale parameter is set according to the coordinate range of the bonding station and the density of bonding points, while the amplitude parameter is calibrated based on the numerical range of historical bonding strength deviation values. The prior parameter is set to a zero-mean prior to match the actual situation in the bonding process where the strength deviation value fluctuates around zero when there is no deviation. The prediction hyperparameters include noise variance and probability confidence interval. The noise variance is set based on the acquisition errors of monitoring sensors and random errors in equipment operation during historical processes. The probability confidence interval is set in conjunction with the yield requirements of the bonding process and the criteria for identifying weak areas. Simultaneously, the preset threshold for the probability distribution of weak strength in the model's output parameters is set in conjunction with the probability of defects such as weak bonding and cold solder joints occurring in actual production and the process compensation requirements.

[0072] Subsequently, the morphological characteristics of the current bonding interface of the first composite material are input into the Gaussian process regression model. Through inference calculation, the probability distribution of the weak strength of the first composite material and the second auxiliary material at multiple bonding sites is obtained, thereby locating the risk areas where problems such as poor bonding, incomplete welding, and delamination may occur.

[0073] Finally, based on the obtained strength weakness probability distribution, the standard bonding parameter matrix is ​​subjected to partitioned weighted compensation. Specifically, the strength weakness probability at each location is obtained through a Gaussian process regression model, and then used as weights to amplify or reduce different coordinate points in the standard bonding parameter matrix, generating a second non-uniform bonding parameter matrix. The standard bonding parameter matrix is ​​a pre-set, uniform table of basic bonding pressure and ultrasonic energy parameters for the entire region, serving as the benchmark for all adaptive compensations. For points with a strength weakness probability greater than a preset threshold, a preset first-step length is used to increase the bonding pressure or ultrasonic energy to enhance the bonding strength. The first-step length L is calculated as: L = k × P × M, where k is the process compensation coefficient (a fixed constant determined by material and bonding process experiments), P is the strength weakness probability (a 0-1 probability value output by the Gaussian process regression model), and M is the average historical bonding strength deviation at that location.

[0074] As can be seen from the above, this embodiment, by retrieving historical bonding quality datasets including actual bonding strength distribution maps and process parameter records, constructs a Gaussian process regression model and predicts the probability distribution of weak bonding based on the current bonding interface morphology characteristics. This enables weighted compensation of the standard bonding parameter matrix, increasing bonding pressure or ultrasonic energy at risk points. This achieves adaptive optimization of secondary bonding parameters based on data-driven and machine learning, effectively reducing the risk of weak areas in multilayer bonding, further improving bonding strength uniformity and product reliability, while giving process parameters the intelligent advantages of self-learning, inheritance, and predictability.

[0075] In one embodiment of this application, before performing the physical bonding of the first composite material identifier and the second auxiliary material, the method further includes: Obtain the first alignment deviation data between the main material and the first auxiliary material in the first composite material, and the second alignment deviation data of the second auxiliary material itself; Based on the first alignment deviation data and the second alignment deviation data, the cumulative alignment error between the second auxiliary material and the main material is calculated. The cumulative alignment error includes X-axis offset, Y-axis offset, rotation angle deviation and warpage deviation. The accumulated alignment error is input to the six-degree-of-freedom motion controller of the bonding platform to generate a compensation motion trajectory. The compensation motion trajectory is used to adjust the spatial attitude of the second auxiliary material during the descent of the bonding head. After the compensation motion trajectory is completed, the bonding head is triggered to perform physical bonding between the second auxiliary material and the first composite material.

[0076] In this embodiment, two sets of alignment deviation data are collected first. For example, one set is the first alignment deviation data that already exists between the main material and the first auxiliary material in the first composite material, and the other set is the positioning, flatness and posture deviation of the second auxiliary material at the current work station, i.e., the second alignment deviation data. These two sets of data together constitute all the error sources before secondary bonding.

[0077] Based on this, the first alignment deviation and the second alignment deviation are fused together to obtain the cumulative alignment error of the second auxiliary material relative to the core main material. This cumulative alignment error includes X-axis offset, Y-axis offset, rotation angle deviation and warping deviation, which represent the relative misalignment, torsion and mismatch of the multilayer materials in space.

[0078] Subsequently, the accumulated alignment error is input to the six-degree-of-freedom motion controller of the bonding platform to obtain the compensated motion trajectory. This compensated trajectory adjusts the spatial attitude of the second auxiliary material or the bonding platform before the bonding head officially descends for bonding, dynamically correcting errors such as offset, rotation, and warping. The six-degree-of-freedom motion controller of the bonding platform is a motion control device that enables six-dimensional spatial attitude adjustment of the bonding platform / second auxiliary material, including translation along the X, Y, and Z axes and rotation around the X, Y, and Z axes. It can plan and execute the compensated motion trajectory based on the accumulated alignment error.

[0079] After the compensation motion trajectory is completed and the attitude correction is in place, the bonding head is triggered to perform physical bonding between the second auxiliary material and the first composite material. This control method of compensation before bonding can significantly improve the alignment accuracy of multilayer bonding, reduce problems such as bonding misalignment, poor soldering, and stress concentration caused by accumulated errors, and effectively improve product yield and structural stability.

[0080] As can be seen from the above, this embodiment obtains the internal alignment deviation of the first composite material and the alignment deviation of the second auxiliary material itself, calculates the cumulative alignment error, and uses a six-degree-of-freedom motion controller to generate a compensating motion trajectory. The spatial attitude is corrected before bonding, which can effectively eliminate the cumulative alignment deviation between the previous process and the current process, further improve the bonding alignment accuracy between the second auxiliary material and the main material, and avoid bonding failure caused by problems such as offset, rotation, and warping.

[0081] In one embodiment of this application, in response to a first debonding signal from the device, the first auxiliary material in the second composite material is debonded, and the debonded composite is flipped to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, a third composite material is obtained, comprising: In response to the device's first debonding signal, the Manufacturing Execution System (MES) client sends a second verification request to the Manufacturing Execution System (MES) server to verify the legality of the debonding operation. The Manufacturing Execution System (MES) server performs verification based on a preset set of debonding verification rules, which includes: current material status verification, process path verification, historical binding relationship verification, batch consistency verification, and physical parameter consistency verification. When all rules in the verification rule set meet the preset debonding conditions, the first auxiliary material in the assembly is debonded, and the debonded assembly is flipped to obtain the third composite material.

[0082] In this embodiment, when the device sends the first debonding signal, it triggers the production execution system client to send a second verification request to the manufacturing execution system server to verify the legality of the debonding operation. The information such as the identity of the combined material to be debonded, the current process, and the equipment status are uploaded in a unified manner, and a strict compliance judgment process is entered to avoid risks such as illegal debonding, unauthorized debonding, and abnormal material status from the source.

[0083] Upon receiving the second verification request, the Manufacturing Execution System (MES) server performs item-by-item verification based on a preset set of debonding verification rules. These rules include verification of the current material status, process path, batch consistency, historical binding relationships, and physical parameter consistency. Specifically, the current material status verification checks whether the second composite material is in a state awaiting debonding. The process path verification confirms that the current equipment is a legitimate debonding station and that the product process has reached a node where debonding is permitted, with no skipped processes, reverse processes, or workstation mismatches. The batch consistency verification checks the batch information of the main material and each auxiliary material in the second composite material, confirming that the relevant batches belong to the same product and the same work order, ensuring batch matching and traceability. The historical binding relationship verification checks the historical binding records corresponding to the material identifiers, confirming that the main and auxiliary materials are in a legitimate state of being bound and not yet unbound, to avoid performing incorrect debonding on unbound or unbound materials. The physical parameter consistency verification confirms that the dimensions, structure, thickness, and other parameters of the composite are consistent with the compatibility requirements of the debonding equipment.

[0084] Only when all the above verification rules are met and pass the verification will the manufacturing execution system server issue a command to the device to allow debonding. At this time, the device initiates the debonding process, safely and non-destructively peeling off the first auxiliary material in the second composite material. After debonding is completed, the remaining composite is flipped over, and the surface pattern of the main material is re-acquired to confirm that it meets the debonding reference plane requirements. Finally, the third composite material is obtained, and the corresponding second composite material identifier is updated synchronously. If any verification fails, the debonding action is stopped and an anomaly is reported.

[0085] As can be seen from the above, this embodiment triggers the legality verification of the debonding operation between the production execution system client and the manufacturing execution system server after the first debonding signal response. It also performs a comprehensive verification based on a set of rules including the current status of materials, process paths, historical binding relationships, batch consistency, and physical parameter consistency. This effectively intercepts problems such as abnormal status, process violations, incorrect binding relationships, batch mismatches, and parameter incompatibility before the debonding process. This avoids material damage and process chaos caused by misunderstanding of bonding, further improves the standardization, security, and traceability of the debonding process, and ensures the stable and reliable operation of the centralized multi-carrier dynamic binding management process for main materials.

[0086] In one embodiment of this application, after debonding the first auxiliary material in the second composite material, flipping the debonded composite to obtain the current surface pattern of the flipped main material, and confirming that it conforms to the debonding reference plane, the third composite material is obtained, and the process further includes: In response to the debonding completion signal, the process execution data collected by the production execution system client during the debonding process is acquired. The production execution system client associates and encapsulates the process execution data with the second combination material identifier, and generates a status update request message; The identifier of the second composite material is parsed based on the status update request message to obtain the identifier information of the main material; Based on the process state after debonding is completed, the material state of the main material is updated from the first process state to the second process state. Remove the historical binding relationship between the identification information of the main material and the identification information of the first auxiliary material and the second auxiliary material, restore the identification information of the main material, and associate and store the identification information with the process execution data to generate a full-process process file for the main material.

[0087] In this embodiment, after the debonding action is completed and a debonding completion signal is triggered, the system enters a closed-loop archiving and intelligent status update stage centered on the main material, to achieve unified management of material status, bonding relationships, and full-process data. The production execution system client will collect process execution data in real time throughout the entire debonding process, including debonding time, equipment parameters, peeling force, debonding results, surface inspection results, posture correction records, and other information.

[0088] Secondly, the production execution system client associates and encapsulates the process execution data with the second composite material identifier, generating a standardized status update request message. This status update request message, indexed by the second composite material identifier, carries all process data of the unbonding stage and is uploaded to the system backend via a stable and reliable communication method.

[0089] Subsequently, the identifiers of the second composite material are parsed, and the core main material identifier information is extracted from them, which is used as a unique index to locate the corresponding material body. After confirming the identity of the main material, the material status of the main material is automatically updated from the previous first process status to the second process status according to the process status after this debonding is completed. Here, the first process status refers to the process status of the main material before debonding (the second composite material has been completed and is waiting to be debonded); the second process status refers to the process status of the main material after debonding is completed (the first auxiliary material has been peeled off and enters the next debonding process).

[0090] Finally, the historical binding relationship is unbound, the association between the main material identifier and the identifiers of the first auxiliary material and the second auxiliary material is unbound, the independent state of the main material identifier is restored, and the process execution data of this debonding is permanently associated with and stored with the main material identifier, generating a full process file of the main material for all stages such as bonding, flipping, correction, and debonding.

[0091] As can be seen from the above, this embodiment improves the accuracy of intelligent unbinding of multi-carrier dynamic binding relationships and material status synchronization by collecting process execution data after debonding, associating and encapsulating it with the second combination material identifier, parsing and updating the main material status, releasing the auxiliary material binding relationship and restoring the main material independent identifier, and storing the entire process data to form a full-process process archive.

[0092] Corresponding to the main material-centralized multi-carrier dynamic binding management method in the above embodiment, Figure 2 This is a structural block diagram of a material-centralized multi-carrier dynamic binding management system provided in one embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The main material-centric multi-carrier dynamic binding management system 20 includes: a material and identification information acquisition module 21, a process surface judgment and flow selection module 22, a first bonding and identification binding module 23, a second bonding and identification combination module 24, a first debonding and main material flipping module 25, and a second debonding and target material acquisition module 26.

[0093] The material and identification information acquisition module 21 is used to acquire the main material, the first auxiliary material, and the second auxiliary material, and to identify the identification information of the main material, the first auxiliary material, and the second auxiliary material. The process surface judgment and process selection module 22 is used to select to enter the bonding process or control the flipping mechanism to perform flipping based on the judgment result of the current surface pattern of the main material and the preset standard process surface. The first bonding and identification binding module 23 is used to respond to the first bonding signal of the device if the bonding process is selected, to physically bond the main material and the first auxiliary material to obtain the first combined material; and to bind the identification information of the main material and the identification information of the first auxiliary material to obtain the identification of the first combined material. The second bonding and identification combination module 24 is used to physically bond the first combined material and the second auxiliary material in response to the second bonding signal of the device to obtain the second combined material, and combine the identification information of the first combined material with the identification information of the second auxiliary material to obtain the identification information of the second combined material. The first debonding and main material flipping module 25 is used to debond the first auxiliary material in the second combined material in response to the first debonding signal of the device, and flip the main material after debonding to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third combined material is obtained. The second debonding and target material acquisition module 26 is used to debond the second auxiliary material in the third combined material in response to the second debonding signal of the device to obtain the target main material.

[0094] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the following modules are described: material and identification information acquisition module 21, process surface judgment and flow selection module 22, first bonding and identification binding module 23, second bonding and identification combination module 24, first debonding and main material flipping module 25, and second debonding and target material acquisition module 26.

[0095] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0096] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0097] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0098] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the main material-centralized multi-carrier dynamic binding management method provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0099] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0100] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0101] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. 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 implementations should not be considered beyond the scope of this application.

[0102] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0104] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0105] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0106] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered 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.

Claims

1. A multi-carrier dynamic binding management method with centralized main material distribution, characterized in that, include: Obtain the main material, the first auxiliary material, and the second auxiliary material, and identify the identification information of the main material, the first auxiliary material, and the second auxiliary material; Based on the judgment result of the current surface pattern of the main material and the preset standard process surface, the process can be selected to enter the bonding process or control the flipping mechanism to perform the flipping. If the bonding process is selected, in response to the first bonding signal of the device, the main material and the first auxiliary material are physically bonded to obtain the first combined material; the identification information of the main material is bound with the identification information of the first auxiliary material to obtain the identification of the first combined material. In response to the second bonding signal of the device, the first composite material and the second auxiliary material are physically bonded to obtain the second composite material, and the identifier of the first composite material is combined with the identifier information of the second auxiliary material to obtain the identifier of the second composite material; In response to the first debonding signal of the device, the first auxiliary material in the second composite material is debonded, and the debonded main material is flipped to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third composite material is obtained. In response to the second debonding signal from the device, the second auxiliary material in the third composite material is debonded to obtain the target main material.

2. The multi-carrier dynamic binding management method for centralized main material management according to claim 1, characterized in that, The determination based on the current surface pattern of the main material and the preset standard process surface, selecting to enter the bonding process or controlling the flipping mechanism to perform flipping, includes: The equipment uses a robotic arm to pick up the main material from the material box and transfer it to the flipping station; The optical sensor on the flipping station acquires the current surface pattern of the main material to obtain the first surface pattern. Based on the identification information of the main material, determine whether the first surface pattern is the standard process surface, and obtain the determination result; If the determination result is that the first surface pattern is a preset standard process surface, then it is bonded to the first auxiliary material; If the determination result is that the first surface pattern is not the standard process surface, then the flipping mechanism is controlled to perform flipping and alignment actions to obtain the second surface pattern.

3. The multi-carrier dynamic binding management method for centralized main material management according to claim 2, characterized in that, If the determination result is that the first surface pattern is not a preset standard process surface, then the flipping mechanism is controlled to perform flipping and alignment actions to obtain the second surface pattern, including: Obtain the deviation angle and / or deviation displacement between the first surface pattern and the standard process surface, and generate a first attitude deviation matrix; The first attitude deviation matrix is ​​input to the motion controller of the flipping mechanism, and the flipping mechanism is driven to perform dynamic pose compensation of the main material through multi-axis linkage during the flipping process to obtain the flipped main material; Image data of the flipped main material is obtained to obtain the second surface map.

4. The multi-carrier dynamic binding management method for centralized main material management according to claim 3, characterized in that, Also includes: Retrieve the reference map corresponding to the preset standard process surface, and perform difference calculation between the second surface map and the reference map to generate a difference image; Identifying particulate contamination or mechanical damage to the host material based on the differential image, and determining bonding pressure parameters or ultrasonic energy parameters for the bonding process based on the particulate contamination or mechanical damage, including: The differential image is segmented at the pixel level to extract the location coordinate set of the particulate contamination or mechanical damage on the main material; The position coordinate set is mapped to the motion trajectory coordinate system of the bonding head to generate a defect spatial distribution map; Based on the defect spatial distribution map, the required compensation bonding energy for each bonding position is calculated, and a non-uniform bonding parameter matrix is ​​generated, wherein the non-uniform bonding parameter matrix includes bonding pressure parameters or ultrasonic energy parameters corresponding to the position coordinate set.

5. The multi-carrier dynamic binding management method for centralized main material management according to claim 4, characterized in that, In response to a first bonding signal from the device, the main material and the first auxiliary material are physically bonded to obtain a first composite material, comprising: In response to the device's first bonding signal, the production execution system client is triggered to send the first verification request for job validity to the production execution system server; The production execution system server receives the first verification request and performs verification based on a preset multi-dimensional verification rule set to obtain the verification result; When the verification result is that all rules in the multi-dimensional verification rule set meet the preset bonding conditions, the main material and the first auxiliary material are physically bonded to obtain the first combined material. If any rule in the multi-dimensional verification rule set does not meet the bonding condition, then the physical bonding between the main material and the first auxiliary material is prevented.

6. The multi-carrier dynamic binding management method for centralized main material management according to claim 5, characterized in that, The physical bonding of the main material and the first auxiliary material to obtain the first combined material includes: Based on the bonding pressure parameter or ultrasonic energy parameter in the non-uniform bonding parameter matrix, the motion controller is activated to apply bonding energy to the region corresponding to the position coordinate set, thereby performing physical bonding with the first auxiliary material. By deploying monitoring sensors in the bonding area, images of the bonding interface or bonding strength characteristic signals of the bonded area are collected. The bonding interface image or bonding strength feature signal is compared with a preset standard bonding quality feature and the deviation is calculated to obtain the quality deviation value. If the quality deviation value is greater than the preset upper limit threshold of the process allowable deviation, then the parameter compensation amount is calculated in reverse based on the quality deviation value as a correction coefficient; The correction coefficient is applied to the remaining unbonded regions in the non-uniform bonding parameter matrix, and the local pressure parameters or local ultrasonic energy parameters of the remaining regions are updated. The bonding head is controlled based on the local pressure parameters or the local ultrasonic energy parameters to continue bonding the remaining areas and obtain the first composite material.

7. The multi-carrier dynamic binding management method for centralized main material management according to claim 1, characterized in that, In response to a second bonding signal from the device, the first composite material and the second auxiliary material are physically bonded to obtain a second composite material. The identifier of the first composite material is combined with the identifier information of the second auxiliary material to obtain a second composite material identifier, including: In response to the second bonding signal from the device, the first bonding process file associated with the first composite material identifier is retrieved; Based on the first bonding process file, the potential weak areas when the first combined material identifier is bonded to the second auxiliary material are predicted, and a second non-uniform bonding parameter matrix for the second auxiliary material is generated. Based on the second non-uniform bonding parameter matrix, the bonding head is controlled to physically bond the first composite material and the second auxiliary material to obtain the second composite material. The identifier of the first composite material and the identifier information of the second auxiliary material are physically combined to obtain the identifier of the second composite material.

8. The multi-carrier dynamic binding management method for centralized main material management according to claim 1, characterized in that, In response to the first debonding signal from the device, the first auxiliary material in the second composite material is debonded, and the debonded composite is flipped to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, a third composite material is obtained, including: In response to the device's first debonding signal, the Manufacturing Execution System (MES) client sends a second verification request to the Manufacturing Execution System (MES) server to verify the legality of the debonding operation. The manufacturing execution system server performs verification based on a preset debonding verification rule set, which includes: current material status verification, process path verification, historical binding relationship verification, batch consistency verification, and physical parameter consistency verification. When all rules in the verification rule set meet the preset debonding conditions, the first auxiliary material in the assembly is debonded, and the debonded assembly is flipped to obtain the third composite material.

9. The multi-carrier dynamic binding management method for centralized main material management according to claim 8, characterized in that, After debonding the first auxiliary material in the second composite material, flipping the debonded assembly, obtaining the current surface pattern of the flipped main material, and confirming that it conforms to the debonding reference plane to obtain the third composite material, the process further includes: In response to the debonding completion signal, based on the process execution data collected by the production execution system client during the debonding process; The production execution system client associates and encapsulates the process execution data with the second combined material identifier to generate a status update request message; Based on the status update request message, the identifier of the second combined material is parsed to obtain the identifier information of the main material; Based on the process state after debonding is completed, the material state of the main material is updated from the first process state to the second process state. Remove the historical binding relationship between the identification information of the main material and the identification information of the first auxiliary material and the identification information of the second auxiliary material, restore the identification information of the main material, and associate and store the identification information with the process execution data to generate a full-process process file for the main material.

10. A multi-carrier dynamic binding management system with centralized main material distribution, characterized in that, include: The material and identification information acquisition module is used to acquire the main material, the first auxiliary material, and the second auxiliary material, and to identify the identification information of the main material, the first auxiliary material, and the second auxiliary material; The process surface judgment and process selection module is used to select to enter the bonding process or control the flipping mechanism to perform flipping based on the judgment result of the current surface pattern of the main material and the preset standard process surface. The first bonding and identification binding module is used to, in response to the first bonding signal of the device, physically bond the main material and the first auxiliary material to obtain the first combined material if the bonding process is selected to enter; and bind the identification information of the main material and the identification information of the first auxiliary material to obtain the identification of the first combined material. The second bonding and identification combination module is used to physically bond the first composite material and the second auxiliary material in response to the second bonding signal of the device to obtain the second composite material, and combine the identification information of the first composite material with the identification information of the second auxiliary material to obtain the identification information of the second composite material. The first debonding and main material flipping module is used to debond the first auxiliary material in the second combined material in response to the first debonding signal of the device, and flip the debonded main material to obtain the current surface pattern of the flipped main material. After confirming that it conforms to the debonding reference plane, the third combined material is obtained. The second debonding and target material acquisition module is used to debond the second auxiliary material in the third combined material in response to the second debonding signal of the device to obtain the target main material.

Images