An automatic transfer and repositioning method and system for electronic product assembly

Abstract

This invention relates to the field of industrial production data management technology, specifically disclosing an automatic transfer and transposition method and system for electronic product assembly. It generates a continuous event flow by collecting and converting production events from each process on the production line. Based on this, it constructs and maintains a graph structure data, defining products, components, workstations, and label carriers as nodes, and their associations as relational edges. The core of this invention lies in its response to labeling requests. The system performs a traversal query based on the graph data. Before the physical labeling action is executed, it dynamically parses the target label carrier information and forcibly verifies the integrity of the product data chain. Only when the verification is successful is a control command generated to drive the actuator to complete the precise transfer, transposition, and labeling of the label carrier. After successful labeling, the system atomically updates the graph data, establishing a binding relationship between the product and the label carrier, achieving forward prediction and reverse full-link traceability. This invention realizes intelligent closed-loop production management from passive response to proactive prediction.

Description

Technical Field

[0001] This invention relates to the field of industrial production data management technology, specifically to an automatic transfer and transposition method and system for electronic product assembly. Background Technology

[0002] In the modern electronics assembly industry, such as for smartphones and tablets, the demand for personalization and customization is increasing, leading to the need to mix and assemble product variations with different configurations on the same production line. After production, these different configurations (such as memory capacity, color, and hardware version) must be assigned a unique identifier (such as a serial number or part number) that strictly corresponds to their actual assembly content to achieve product identification, management, and full lifecycle traceability. One approach is to manually select and affix the corresponding physical identifier based on the product configuration information. This method is inefficient and prone to human error, resulting in mismatches between identifiers and products. Another approach is to use semi-automatic equipment that scans the product and retrieves the identifier information from a pre-set static list for physical attachment.

[0003] The fundamental flaw in these existing methods lies in their reliance on a static or predefined configuration list for execution logic. They fail to provide real-time, dynamic verification of whether the "product entity" about to be assigned an identifier maintains consistency with its digital "data identity" throughout the entire production chain. In other words, there is a risk of disconnect and asynchrony between the physical assembly flow (physical flow) and the accompanying process data and component data flows (information flow). This can lead to physical identifiers being bound to an incomplete or misconfigured online product, thus contaminating the entire production data chain at its source and compromising the accuracy and reliability of subsequent management processes such as quality traceability and batch recalls. Summary of the Invention

[0004] The purpose of this invention is to provide an automated transfer and transposition method and system for assembling electronic products, in order to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] An automated transfer and transposition method for assembling electronic products includes the following steps:

[0007] S1: Collect and transform production events from multiple processes on the electronic product assembly line to generate continuous production event stream data;

[0008] S2: Based on production event flow data, establish and maintain a graph structure data, in which products, components, process stations and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges;

[0009] S3: Perform graph traversal query based on graph structure data, parse out the target identification carrier information that the current product needs to be bound, and verify whether the data chain related to the current product is complete before the attachment operation is executed;

[0010] S4: Only when the data chain integrity verification is passed, generate and execute labeling control instructions to drive the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier;

[0011] S5: After confirming that the target identification carrier has been correctly attached, submit and update the graph structure data as a new relationship edge representing the successful binding relationship between the product and the identification carrier;

[0012] S6: Based on continuously updated graph structure data, it enables positive prediction and reverse full-link traceability of data flow for online products;

[0013] Among them, the identification carrier refers to any information that carries a unique identity information bound to an electronic product (such as serial number, part number, configuration code).

[0014] As a further aspect of the present invention, the specific process of S2 is as follows:

[0015] Real-time analysis of continuously incoming production event stream data is performed to extract event subject identifiers, event types, and event association attributes;

[0016] Based on the event type, match the predefined graph operation rules and execute the corresponding node and relationship edge operations: when the event type is process completion, establish an experience relationship edge between the product node and the process station node; when the event type is configuration lock, establish an inclusion relationship edge between the product node and the component node.

[0017] In response to the trigger signal of the labeling station, starting from the current product node, a bidirectional traversal is performed along the experience relationship edge and the containment relationship edge to verify the integrity of the path to the current node, and extract all the identification carrier node information associated with the end component node as the target identification carrier information.

[0018] As a further aspect of the present invention: the response to the trigger signal at the labeling station, starting from the current product node, involves a bidirectional traversal along the edges of the experience relationship and the inclusion relationship, specifically including:

[0019] Define and obtain the reverse tracing depth centered on the current product node and along the experience relationship edge; dynamically set the time window based on the product process specification; and filter all process station nodes created within the time window that are directly related to the current product node.

[0020] Verify that each process station node has a status attribute that represents the process being qualified, and confirm that all necessary process station nodes exist in the established associated paths, thereby verifying the integrity of the path to the current node.

[0021] If and only if the path integrity verification passes, start from the current product node, traverse along the containment relationship edge to all end component nodes, and aggregate all the identifier carrier node information connected by the configuration relationship edge on each end component node to generate the final target identifier carrier information set.

[0022] As a further aspect of the present invention, the specific process of S3 is as follows:

[0023] Obtain all established relationship edges of the current product node in the graph structure data to form a complete set of associated paths from the product's initial process to the current labeling station;

[0024] The associated path set is compared with the predefined process route template to verify whether all key process station nodes are included in the associated path and whether the connection order of the relationship edges between each process station node conforms to the specification requirements of the process route template.

[0025] Check the status attributes of each process station node in the associated path set, confirm that the status attributes of all necessary process station nodes are marked as qualified, and identify process station nodes with missing status or abnormal markings.

[0026] The data chain integrity verification is deemed successful only if the set of associated paths completely covers the process route template and all key node statuses are qualified. Based on the binding relationship edges between component nodes and identification carrier nodes, the final target identification carrier information set is parsed and generated.

[0027] As a further aspect of the present invention: obtaining all established relation edges of the current product node in the graph structure data to form a complete set of associated paths from the initial process to the current labeling station specifically includes:

[0028] Starting from the current product node, perform multi-level reverse traversal along the experience relationship edge to dynamically establish a complete process path from the current labeling station node back to the starting process node;

[0029] In the complete process path, based on the event timestamp attribute recorded by each process station node, the temporal continuity of each relation edge is verified to ensure that the order in which the product flows through each process station meets the predetermined production cycle requirements.

[0030] Synchronously traverse the containment edges in the graph structure data, extract all component nodes associated with the current product node, and establish a mapping relationship between component nodes and corresponding process path nodes;

[0031] The complete process path and component mapping relationship that have been verified by time sequence are integrated to generate a multi-dimensional set of related paths containing time dimension, process dimension and component dimension, which serves as the input basis for data chain integrity verification.

[0032] As a further aspect of the present invention, the specific process of S4 is as follows:

[0033] Based on the target identification carrier information set generated after the data chain integrity verification is passed, the physical attribute parameters of each target identification carrier are analyzed, including the identification carrier size, material type and coordinate position in the feeder.

[0034] A set of multi-dimensional control parameters, including transfer trajectory, transposition angle and attachment pressure, is dynamically generated based on physical property parameters.

[0035] Based on a set of multi-dimensional control parameters, segmented motion commands are generated to drive the actuator to sequentially complete the following tasks: accurately picking up the label carrier from the feeder, transposing and adjusting it in three-dimensional space according to a specified trajectory and angle, and accurately attaching the label carrier to the specified area of ​​the product with optimized attachment pressure.

[0036] During the continuous contact between the label carrier and the product surface, the matching degree between the adhesion pressure curve and the preset ideal pressure curve is analyzed in real time, and the end posture of the actuator is adjusted.

[0037] As a further aspect of the present invention: the precise application of the label carrier to the designated area of ​​the product specifically includes:

[0038] Based on the physical properties of the label carrier and the target attachment position, plan the continuous motion trajectory sequence of the end effector of the actuator from the feeder to the product surface. The continuous motion trajectory sequence includes the pickup preparation section, the label carrier pickup section, the spatial transposition section, and the attachment approach section.

[0039] In the tag carrier pickup section, the pickup posture of the end effector is dynamically adjusted based on the peeling characteristics of the tag carrier material;

[0040] In the spatial transposition segment, based on the current actual spatial posture of the product and the spatial geometric relationship between the target attachment surface of the marker carrier, the Euler angle transformation sequence of the end effector in three-dimensional space is calculated in real time to achieve a smooth transposition of the marker carrier posture.

[0041] In the approach segment, based on the microscopic morphological features of the product surface fed back by the visual positioning system, the approach path and application pressure curve of the end effector are optimized to achieve progressive contact and complete adhesion between the label carrier and the product surface.

[0042] As a further aspect of the present invention, the specific process of S5 is as follows:

[0043] Collect multi-dimensional image data of the attached label carriers, and comprehensively analyze the positional accuracy, angle deviation and adhesion integrity of the label carriers based on the preset label carrier adhesion quality evaluation rules to generate adhesion verification results;

[0044] When the attachment verification result meets the preset standard, extract the unique identifier of the current product node and the serial number information of the target identifier carrier node, and construct a binding data packet containing timestamp and workstation information;

[0045] The binding relationship data packet is submitted to the graph structure database as an atomic operation transaction. A binding relationship edge with complete attributes is established between the product node and the identifier carrier node, and the status attribute of the product node is updated to "labeled" synchronously.

[0046] Based on the newly established binding relationship edge-to-graph structure data, a consistency check is performed to verify the uniqueness and consistency of the correspondence between product nodes and identifier carrier nodes throughout the entire production path.

[0047] As a further aspect of the present invention, the specific process of S6 is as follows:

[0048] Real-time monitoring of newly added binding relationship edges in the graph structure data; starting from the product node corresponding to the newly added binding relationship edge, traversing the subsequent process path in the forward direction along the relationship edge; preloading the configuration parameters required by the downstream process and performing resource availability verification in advance.

[0049] When an abnormal change in the status attribute of any process station node is detected, the corresponding abnormal node is used as the base point, and all related product nodes are traversed in reverse along the experience relationship edge to construct a complete traceability chain for the affected product.

[0050] Based on the temporal attributes and topological relationships of each node in the complete traceability chain, a disposal plan including batch isolation suggestions and reprocessing paths is generated, and the disposal status of the affected product nodes is marked in the graph structure data.

[0051] By continuously analyzing the frequency and distribution patterns of various relationship edges in graph structure data, bottleneck process paths in the production process can be identified.

[0052] An automated transfer and transposition system for assembling electronic products includes:

[0053] The production event stream acquisition and conversion module is used to collect and convert production events from multiple processes on the electronic product assembly line to generate continuous production event stream data.

[0054] The graph structure data construction and maintenance module establishes and maintains a graph structure data based on production event flow data. In this module, products, components, process stations, and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges.

[0055] The graph traversal query and verification module performs graph traversal queries based on graph structure data, parses out the target identifier carrier information that the current product needs to be bound, and verifies whether the data chain related to the current product is complete before the attachment operation is performed.

[0056] The labeling control instruction generation and execution module generates and executes labeling control instructions only when the data link integrity verification is passed, driving the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier;

[0057] The binding relationship update module, after confirming that the target identification carrier has been correctly attached, submits and updates the binding relationship that represents the successful binding between the product and the identification carrier as a new relationship edge in the graph structure data.

[0058] The data flow analysis and traceability module, based on continuously updated graph structure data, enables forward prediction and reverse full-link traceability of data flow for online products.

[0059] The beneficial effects of this invention are:

[0060] (1) This invention constructs real-time updated graph structure data and performs data chain integrity verification based on graph traversal before physical attachment operations, ensuring absolute synchronization between information flow and physical flow. This not only prevents erroneous binding at the source, but also makes the full life cycle data of each product (from components to processes, and then to identification carriers) form a complete and reliable "digital lineage", providing unprecedented data credibility and integrity assurance for quality traceability, root cause analysis of problems, and precise recall.

[0061] (2) This invention uses continuously updated graph structure data to dynamically and positively predict and verify the resource requirements of downstream processes in advance. Simultaneously, it can perform reverse full-link tracing of any node anomalies, quickly constructing an impact range map. This makes the production process no longer a passive response to problems, but rather an active prediction of bottlenecks, proactive risk avoidance, and rapid location of the root cause and precise isolation of affected batches when problems occur, automatically generating the optimal reprocessing path. Thus, the single process of transferring the identification carrier is elevated to a core link driving dynamic optimization and intelligent decision-making across the entire production line, improving production flexibility and overall efficiency. Attached Figure Description

[0062] The invention will now be further described with reference to the accompanying drawings.

[0063] Figure 1This is a flowchart of the method of the present invention;

[0064] Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Please see Figure 1 As shown, this invention provides an automated transfer and transposition method for electronic product assembly, comprising the following steps:

[0067] S1: Collect and transform production events from multiple processes on the electronic product assembly line to generate continuous production event stream data;

[0068] S2: Based on production event flow data, establish and maintain a graph structure data, in which products, components, process stations and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges;

[0069] S3: Perform graph traversal query based on graph structure data, parse out the target identification carrier information that the current product needs to be bound, and verify whether the data chain related to the current product is complete before the attachment operation is executed;

[0070] S4: Only when the data chain integrity verification is passed, generate and execute labeling control instructions to drive the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier;

[0071] S5: After confirming that the target identification carrier has been correctly attached, submit and update the graph structure data as a new relationship edge representing the successful binding relationship between the product and the identification carrier;

[0072] S6: Based on continuously updated graph structure data, it enables forward prediction and reverse full-link traceability of data flow for online products.

[0073] In S1, production events from multiple processes on the electronic product assembly line are collected and transformed to generate continuous production event stream data, specifically including:

[0074] During the programming process, a programming completion event is acquired through the communication interface of the programming device. This event data includes the product's unique identifier and the software version information being programmed. After completing the program writing, the programming device automatically generates a completion signal containing a timestamp, which is transmitted to the data processing unit through the device's built-in data acquisition interface.

[0075] During the testing process, the output interface of the testing instrument captures test result events. This event data includes the product serial number, test item number, and test pass / fail status. After completing each performance test, the testing instrument sends the structured test result data to the data aggregation point via a standard communication protocol.

[0076] During the assembly process, displacement and torque sensors mounted on the robotic arm collect data on component assembly completion events. This event data includes the component model, assembly force curve, and assembly position coordinates. The analog signals collected by the sensors are converted into digital signals by an analog-to-digital converter before being uploaded.

[0077] In the labeling station triggering process, photoelectric sensors and proximity switches are used to detect the product's arrival signal, which includes the pallet identification number and arrival time. When the product enters the labeling station detection area, the photoelectric sensor generates a level change signal, which is converted into processable event data through a digital input module.

[0078] In the environmental monitoring phase, temperature and humidity sensors collect real-time parameters of the production environment. This data includes measured values, collection locations, and collection times. The sensors convert the measured physical quantities into digital quantities in standard units for transmission via an industrial bus.

[0079] All collected event data undergoes format standardization processing, uniformly converting it into a standard data structure that includes event type, occurrence time, equipment identifier, and data payload, forming a continuous production event stream data.

[0080] In S2, a graph structure is established and maintained based on production event flow data. Products, components, process stations, and identifiers are defined as nodes, and the relationships between nodes are defined as edges. Specifically, these include:

[0081] When performing real-time analysis of continuously incoming production event stream data, a data analysis script is used to extract key elements from the events. This analysis process separates three structured fields from the event data: an event subject identifier field to identify the product number or equipment number; an event type field to distinguish business types such as programming completion, testing passed, and assembly completed; and an event association attribute field to extract specific business data such as software version number, test values, and assembly parameters. This parsed structured data is organized into records in a unified format for subsequent graph construction.

[0082] When matching predefined graph operation rules based on event types, a rule mapping table is used to establish the correspondence between event types and graph operations. This rule mapping table contains two columns of configuration information: the first column is the event type code, and the second column is the corresponding graph operation instruction. When an event of the "process completion" type is received, the operation of creating an experience relationship edge is executed. Specifically, this operation executes a creation statement in the graph database, establishing a connection edge with a timestamp attribute between the product node and the process station node. When an event of the "configuration lock" type is received, the operation of creating a containing relationship edge is executed. This operation establishes a connection edge with configuration parameter attributes between the product node and the component node.

[0083] When performing a bidirectional traversal in response to a trigger signal from the labeling station, the traversal depth is first determined. The traversal depth is determined by counting the number of traversed relational edges backwards from the current product node. The traversal stops when the count reaches the minimum number of processes defined in the product process specification. The time window is set based on the maximum allowable production cycle specified in the product process specification. The starting boundary of the time window is obtained by subtracting the maximum allowable production cycle from the current time. Based on these two parameters, a set of process station nodes created within the time window and directly associated with the current product node is selected.

[0084] The process of verifying the completeness of the process path involves two verification steps. The first step checks whether each process station node possesses a status attribute indicating that the process is qualified. The value of this status attribute is derived from the actual result data recorded in the test process. The second verification step compares the set of associated process station nodes with the list of required processes defined in the product process specification. This list of required processes is derived from the process route definition in the product engineering documents. The results of these two verification steps are used to comprehensively determine whether the path is complete.

[0085] The operation to generate the target identifier carrier information set is performed after the path integrity verification is passed. This operation starts from the current product node and traverses along the containment edges to all end component nodes. The identification criterion for an end component node is that it has no subsequent containment edges. Then, it collects information on all identifier carrier nodes connected to these end component nodes through configuration relationship edges, which are static relationships pre-established during component specification definition. Finally, the collected identifier carrier node information is organized into a structured data set according to component categories.

[0086] In S3, a graph traversal query is performed based on the graph structure data to parse out the target identifier carrier information that the current product needs to be bound to. Before the attachment operation is executed, the integrity of the data chain related to the current product is verified, specifically including:

[0087] When retrieving all established relationship edges for the current product node in the graph structure data, a depth-first search algorithm is used for multi-level reverse traversal. This traversal process starts with the current product node and backtracks layer by layer along the reverse direction of the traversed relationship edges, which are connections established during the process completion event handling. During the traversal, all process station nodes visited and their connection order are recorded until the starting process node is reached. The starting process node is defined as a process station node without any predecessor traversed relationship edges. Through this traversal method, a complete set of associated paths from the starting process to the current labeling station is formed. This set contains the complete topology of all nodes in the path and their connections.

[0088] When verifying the temporal continuity within a complete process path, a time series analysis method is employed. This method first extracts specific time values ​​from the event timestamp attributes recorded at each process station node. These timestamp attributes represent time information recorded during the processing of production event flow data. Then, following the node order in the process path, the difference in timestamps between adjacent nodes is calculated to obtain the actual time interval between each process step. These actual time intervals are then compared with predetermined production cycle time requirements, which are derived from the standard time consumption of each process step defined in the product process specification document. By calculating the absolute value of the deviation between the actual time interval and the standard time consumption, the temporal continuity is determined to meet the requirements. The absolute value of the deviation is calculated by subtracting the standard time consumption from the actual time interval and taking the absolute value.

[0089] When synchronously traversing containment edges, a breadth-first search algorithm is used to extract component node information. This traversal process starts from the current product node and visits all directly associated component nodes along the first-level connections of the containment edges. The containment edges are connections established during the configuration lock event handling process. For each visited component node, its correspondence with the source process path node is recorded, establishing a mapping table between component nodes and process path nodes. This mapping table contains three columns: the component node identifier, the corresponding process path node identifier, and the timestamp when the relationship was established.

[0090] When integrating process paths with component mapping relationships, data fusion technology is used to generate a multi-dimensional set of associated paths. This process first uses the node sequence in the process path as time-dimensional data, recording the execution order of each process. Then, the node attributes in the process path are used as process-dimensional data, containing the technical parameters and quality requirements of each process. Finally, information from the component mapping relationship table is used as component-dimensional data, containing the specifications and configuration information of each component. These three dimensions of data are linked and fused through a common product node identifier to form a set of associated paths containing complete production information.

[0091] When comparing the set of associated paths with the predefined process route template, a graph pattern matching method is used. The predefined process route template is stored in the form of a standard process route graph, containing a set of critical process station nodes that must be included and their prescribed connection order. The comparison process first checks whether the set of associated paths contains all the critical process station nodes defined in the process route template. Critical process station nodes are necessary process nodes determined according to product process requirements. Then, it verifies whether the connection order of these critical nodes in the set of associated paths is consistent with the order defined in the process route template. The consistency of the connection order is determined by comparing the edge directions between adjacent nodes.

[0092] When checking the status attributes of process station nodes, a state machine verification mechanism is used. This verification process first retrieves the status attribute value of each process station node from the associated path set. The status attribute value is the pass or failure status recorded when processing test result events. Then, based on the list of required processes defined in the process route template, all process station nodes that must be checked are selected. For each required node, its status attribute value is checked to see if it is marked as qualified. The specific value of the qualified status comes from the pass status code defined in the quality inspection standard. Simultaneously, nodes with empty status attribute values ​​or belonging to abnormal status codes are identified, generating an abnormal node list.

[0093] When determining the integrity of the data chain, a multi-condition judgment logic is employed. This judgment process first requires that the set of associated paths completely cover the process route template; that is, all critical process station nodes in the template must exist in the associated paths, and the connection order must be completely consistent. Secondly, the status attributes of all critical process station nodes must be marked as qualified, with no abnormal or missing statuses. Both conditions must be met simultaneously for the data chain integrity verification to pass. If either condition is not met, the process is immediately terminated, and a corresponding error report is generated.

[0094] When parsing and generating the target identification carrier information set, a graph traversal query method is used. This method is based on the pre-established binding relationship edges between component nodes and identification carrier nodes. These binding relationship edges represent the correspondence between components and identification carriers established during material management. The traversal process starts from all component nodes included in the current product node, visits the corresponding identification carrier nodes along the binding relationship edges, and collects information on all reachable identification carrier nodes. Then, this identification carrier node information is organized according to component categories to generate a structured target identification carrier information set. This set contains the identifier, specifications, and corresponding component information for each identification carrier.

[0095] In the specific process of verifying the continuity of the time sequence, the time deviation threshold is set based on the allowable deviation range in the product process specifications. This threshold is obtained by multiplying the standard time consumption by a percentage factor, which is adjusted between 5% and 10% according to the criticality of the process. A lower percentage factor is used for critical processes, and a higher percentage factor is used for non-critical processes. In actual calculation, the actual time interval of each process is compared with the standard time consumption. If the absolute value of the deviation exceeds the calculated time deviation threshold, the time sequence of that process is determined to be abnormal.

[0096] When checking status attributes, the specific criteria for judging a qualified status are based on the definitions in the quality inspection procedures. A qualified status must simultaneously meet three conditions: the status attribute value must exist in a predefined list of valid status values, the timestamp of the status record must be within the product's production cycle, and the equipment identifier of the status record must be authorized and certified. Verification of these conditions is achieved by querying valid status records in the quality inspection database, ensuring the authenticity and validity of the status information.

[0097] In S4, labeling control instructions are generated and executed only when the data link integrity verification passes, driving the automated actuator to complete the transfer, spatial transposition, and attachment operations of the target label carrier, specifically including:

[0098] Based on the target identification carrier information set generated after the data chain integrity verification is passed, the process of parsing the physical attribute parameters of each target identification carrier is as follows: The identification carrier size parameters, including length and width values, are extracted from the identification carrier information database. The length value is obtained by measuring the actual outer contour dimension of the identification carrier on the base paper, and the width value is obtained by measuring the dimension perpendicular to the length direction. The material type is determined by querying the material classification table corresponding to the identification carrier's material code. This classification table includes physical characteristic parameters such as material hardness coefficient and surface friction coefficient. The coordinate position is obtained by reading the value of the position encoder on the feeder. This value indicates the specific position of the identification carrier on the feeder's conveyor belt. The position encoder reading is then converted to obtain the three-dimensional coordinate value in the machine coordinate system.

[0099] The specific calculation process for dynamically generating a multi-dimensional control parameter set based on physical property parameters is as follows: The transfer trajectory is planned using a cubic spline interpolation algorithm. A smooth motion path is generated by setting the coordinates of the trajectory's starting point, ending point, and intermediate control points. The starting point coordinates are the coordinates of the marking carrier on the feeder, and the ending point coordinates are the coordinates of the product attachment position. The transpose angle is calculated using a spatial coordinate system transformation method. The required rotation angle is determined by calculating the angle between the normal vector of the product attachment surface and the initial normal vector of the marking carrier. This angle is decomposed into three components: rotation angle around the X-axis, rotation angle around the Y-axis, and rotation angle around the Z-axis. The attachment pressure is calculated according to the formula... ,in Indicates the adhesion pressure value. Indicates the area of ​​the sign carrier. Indicates the coefficient of friction of the surface of the label carrier material. Indicates the adhesion speed. , , These are the area influence coefficient, friction influence coefficient, and velocity influence coefficient, which are determined through materials mechanics experiments.

[0100] The specific implementation process of generating segmented motion commands based on a multi-dimensional control parameter set is as follows: In the pre-pickup segment, the motion command-controlled end effector moves from the waiting position to a certain height above the label carrier pickup position. This height is determined based on the label carrier size, typically set to 1.2 times the label carrier length. In the label carrier pickup segment, the motion command-controlled end effector descends vertically and, in conjunction with a vacuum suction cup, picks up the label carrier. The descent speed is adjusted according to the label carrier material; a lower speed is used for more brittle materials. In the spatial transposition segment, the motion command-controlled end effector synchronously adjusts its angle during movement. The angle adjustment uses a uniform speed change to ensure all angle adjustments are completed before reaching the attachment position. In the attachment approach segment, the motion command-controlled end effector presses the label carrier onto the product surface with a preset attachment pressure. The pressure value is adjusted in real-time based on the previously calculated attachment pressure value.

[0101] The specific method for real-time monitoring during the contact between the label carrier and the product surface is as follows: Actual attachment pressure data is collected using a pressure sensor installed on the end effector. The sampling interval of the pressure sensor is 10 milliseconds, and the collected pressure data forms a real-time pressure curve. The real-time pressure curve is then compared with a preset ideal pressure curve for similarity calculation. The root mean square error (RMSE) method is used for the similarity calculation, and the formula is: ,in This represents the root mean square error. Indicates the first The actual pressure value at each sampling point Indicates the first Ideal pressure value at each sampling point This represents the total number of sampling points. When the root mean square error exceeds the threshold, the attitude of the end effector is adjusted, and the adjustment amount is determined based on the magnitude and direction of the error.

[0102] The specific process for planning the continuous motion trajectory sequence is as follows: The trajectory planning for the pre-pickup segment uses a linear interpolation algorithm, moving linearly from the current position to a safe height above the marker carrier. The trajectory planning for the marker carrier pickup segment uses a vertically descending linear motion, switching to a low-speed mode when 1 mm from the marker carrier surface. The trajectory planning for the spatial transposition segment uses a spatial circular interpolation algorithm to ensure the end effector maintains stable movement during transposition. The trajectory planning for the attachment approach segment uses linear motion combined with pressure control, initiating pressure control mode when 0.5 mm from the product surface.

[0103] The specific implementation methods for adjusting the pickup posture based on the peeling characteristics of the label carrier material are as follows: For highly viscous label carriers, a slight lateral movement is added during pickup, with the movement distance being twice the thickness of the label carrier, to help overcome the initial adhesion force. For brittle label carriers, a segmented depressurization suction method is adopted, first contacting the label carrier surface with a lower vacuum pressure, and then increasing to the standard vacuum pressure after the contact is stable. For ultra-thin label carriers, a slight tilt angle is maintained during pickup, the size of which is calculated according to the size of the label carrier and is usually controlled between 0.5 degrees and 1 degree.

[0104] The calculation process for real-time Euler angle transformation sequence is as follows: First, the actual spatial posture data of the product, including the normal vector coordinates of the product's attachment surface, is acquired through a vision sensor installed on the production line. Then, the transformation matrix from the current orientation of the marker carrier to the target orientation is calculated. This transformation matrix is ​​obtained through a basis transformation between the two coordinate systems. The transformation matrix is ​​decomposed into Euler angle form, yielding the rotation angles α around the X-axis, β around the Y-axis, and γ around the Z-axis. During the transpose process, the rotation operations are performed step-by-step in the order of α, β, and γ, with the time interval for each rotation step evenly distributed according to the total time of the transpose segment.

[0105] The specific method for optimizing the approach path and adhesion pressure curve based on visual positioning system feedback is as follows: A high-resolution industrial camera is used to acquire microscopic images of the product surface. The images are then processed for grayscale and edge detection to identify surface unevenness. The approach path is adjusted according to the surface undulation; for raised areas, the end effector height is increased accordingly. The adjustment amount is calculated using the following formula: ,in, Indicates the adjustment amount. Indicates the height of the protrusion. Let the safety factor be 0.8. At the same time, the adhesion pressure curve is adjusted according to the surface roughness. A higher initial adhesion pressure is adopted for surfaces with larger roughness, and the pressure adjustment amount is determined based on a pre-established surface roughness-pressure correspondence table. During the adhesion process, the contact state between the identification carrier and the surface is continuously monitored, and the pressure value is fine-tuned in real time through the feedback of the pressure sensor to ensure that the identification carrier is fully adhered to the surface.

[0106] In S5, after confirming that the target identification carrier is correctly adhered, the binding relationship representing the successful binding of the product and the identification carrier is submitted and updated as a new relationship edge to the graph structure data, specifically including:

[0107] When collecting multi-dimensional image data of the adhered identification carrier, 3 industrial cameras installed above and on the side of the labeling station are used to obtain images. The upper camera shoots vertically downward to obtain the overall position of the identification carrier, and the 2 side cameras shoot at a 45-degree angle to detect the fitting state of the edge of the identification carrier. These cameras cooperate with the ring light source and the coaxial light source to eliminate reflection and highlight the surface texture respectively. The collected color images are converted into grayscale images, and then the contour features of the identification carrier are identified through edge detection algorithms. The calculation of the position accuracy is by measuring the straight-line distance between the center point coordinates of the contour of the identification carrier and the center point coordinates of the preset target adhesion area on the product, and this distance value is the position deviation value. The calculation of the angle deviation is obtained by comparing the included angle value between the main direction axis of the contour of the identification carrier and the standard direction axis of the target adhesion area. The analysis of the fitting integrity is to evaluate by calculating the percentage of the non-fitting area between the contour area of the identification carrier and the product surface in the total area.

[0108] When analyzing based on the preset evaluation rules for the adhesion quality of the identification carrier, the set position accuracy threshold is 0.5 mm, the angle deviation threshold is 0.5 degrees, and the fitting integrity threshold requires that the percentage of the non-fitting area is less than 1%. The evaluation process adopts a hierarchical judgment logic. First, check whether the position accuracy is within the threshold range, then check whether the angle deviation meets the requirements, and finally verify whether the fitting integrity meets the standard. Only when all these 3 indicators meet the requirements, the adhesion verification result is judged to be qualified. If any indicator exceeds the threshold range, record the specific out-of-tolerance items and values, and generate an unqualified verification result.

[0109] When the labeling verification result meets the preset standards, the binding relationship data package construction operation is performed. The unique identifier of the current product node is extracted from the graph structure data; this identifier is a 16-digit string consisting of numbers and letters. Simultaneously, the identification carrier serial number is extracted from the target identification carrier information; this serial number is a 20-digit string containing the production batch and a unique number. The binding relationship data package contains five fields: product identifier field, identification carrier serial number field, timestamp field, workstation number field, and verification result field. The timestamp field records the precise time of labeling completion, accurate to the millisecond level. The workstation number field records the physical workstation identifier where the labeling operation was performed. The verification result field stores the specific test values ​​and quality level.

[0110] When committing the binding relationship data packet as an atomic operation transaction, a database transaction management mechanism is employed to ensure the integrity of the operation. The process first sends a start transaction command to the graph database, then executes the operation to create a binding relationship edge, establishing a new connection edge between the product node and the identifier carrier node. This binding relationship edge contains five attributes: binding time, operation location, verification status, position deviation value, and angle deviation value. Immediately after successfully creating the relationship edge, the operation to update the product node's status attribute is executed, changing the product node's status from pending labeling to labeled. These two operations are completed within a single transaction, ensuring either complete success or complete rollback to avoid data inconsistencies.

[0111] When performing consistency checks based on newly established binding edges, a graph traversal query method is used to verify the uniqueness of the relationships. This query starts with the newly established binding edge and checks whether other product nodes in the entire production path have established binding relationships with the same identifier carrier serial number. It also checks whether the current product node already has other valid binding edges. Consistency checks also include verifying the reasonableness of the time sequence of binding relationships, ensuring that the labeling operation time is later than the completion time of all preceding processes and earlier than the start time of all subsequent processes. For detected inconsistencies, detailed inconsistency information is recorded, and a data repair process is triggered to ensure that the correspondence between products and identifier carriers in the graph structure data remains correct and consistent.

[0112] In the positional accuracy calculation process, the specific calculation of the straight-line distance adopts the distance formula between two points in a two-dimensional coordinate system. First, a Cartesian coordinate system is established with the center of the product target attachment area as the origin. The coordinate values ​​of the center point of the marker carrier outline are obtained through image processing, represented by x and y values ​​respectively. The distance value is calculated by first adding the squares of the difference between the x and y coordinates, and then taking the square root of the sum. The resulting value is the positional deviation value. This calculation process is implemented through the geometric measurement function in the image processing software to ensure the accuracy of the measurement results.

[0113] In S6, based on continuously updated graph-structured data, forward prediction and reverse end-to-end tracing of data flow for online products are achieved, specifically including:

[0114] When new binding edges are added to the graph structure data in real time, an event listening mechanism is used to capture data changes. When a new binding edge is detected, a breadth-first search algorithm is immediately used to traverse forward along the edge, starting from the corresponding product node. The traversal depth is set to 3 nodes, covering the 3 consecutive process nodes following the current process. Preloading is implemented by querying the process parameter database to obtain the equipment configuration parameters, bill of materials parameters, and quality standard parameters required by downstream processes. Resource availability verification is completed by comparing equipment status data with process requirement parameters. Equipment status data comes from real-time operating data collected by equipment monitoring sensors, including equipment operating status, current load, and estimated idle time. During the verification process, equipment utilization is calculated; when equipment utilization exceeds 95%, a resource shortage state is determined.

[0115] When an abnormal change in the status attribute of a process station node is detected, a reverse tracing process is initiated. Abnormal status identification is based on mutation detection of status attribute values; tracing is triggered when the status value changes from a normal status code to an abnormal status code. Using the abnormal node as a base point, a depth-first search algorithm is employed to traverse backwards along the traversal edges, covering all directly or indirectly related product nodes. The constructed complete traceability chain contains five key elements: product node identifier, process path sequence, timestamp sequence, status value sequence, and operator identifier. The traceability chain is stored using a chained data structure, with each node storing references to its predecessor node.

[0116] When generating a disposal plan based on a complete traceability chain, the temporal attributes of each node in the chain are first analyzed. Temporal analysis includes calculating process intervals, anomaly occurrence times, and duration of impact. Topology analysis primarily examines the connection density and path length between nodes. The disposal plan is generated using a rule-based reasoning method; the rule base contains 20 disposal rules covering responses to different anomaly types and impact ranges. Batch isolation recommendations are generated based on the spatial location data of product nodes and process similarity calculations, clustering affected product nodes into 2 to 5 isolation batches according to process characteristics. Reprocessing path planning is achieved by recalculating the process route, avoiding anomalous process nodes, and selecting alternative process paths.

[0117] When identifying bottleneck process paths through continuous analysis of graph structure data, the statistical period is set to 8 hours. The analysis process primarily monitors three key indicators: the variance of the time interval for establishing relationship edges, the average node waiting time, and the frequency of resource conflicts. The frequency of relationship edge establishment is calculated using a sliding window statistical method, with a window size set to 100 production cycles. Distribution pattern analysis employs a clustering algorithm, categorizing process paths into three levels based on execution efficiency. Bottleneck process paths are identified by simultaneously meeting three conditions: the variance of the time interval for establishing relationship edges is greater than a set threshold, the average node waiting time exceeds 1.5 times the standard process time, and the frequency of resource conflicts reaches more than 5 times within the statistical period. Identified bottleneck paths are marked with priority levels, providing data for production scheduling optimization.

[0118] Please see Figure 2 As shown, an automated transfer and transposition system for assembling electronic products includes:

[0119] The production event stream acquisition and conversion module is used to collect and convert production events from multiple processes on the electronic product assembly line to generate continuous production event stream data.

[0120] The graph structure data construction and maintenance module establishes and maintains a graph structure data based on production event flow data. In this module, products, components, process stations, and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges.

[0121] The graph traversal query and verification module performs graph traversal queries based on graph structure data, parses out the target identifier carrier information that the current product needs to be bound, and verifies whether the data chain related to the current product is complete before the attachment operation is performed.

[0122] The labeling control instruction generation and execution module generates and executes labeling control instructions only when the data link integrity verification is passed, driving the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier;

[0123] The binding relationship update module, after confirming that the target identification carrier has been correctly attached, submits and updates the binding relationship that represents the successful binding between the product and the identification carrier as a new relationship edge in the graph structure data.

[0124] The data flow analysis and traceability module, based on continuously updated graph structure data, enables forward prediction and reverse full-link traceability of data flow for online products.

[0125] The working principle of this invention is as follows: First, production events from processes such as programming, testing, and assembly are collected and converted to generate continuous production event stream data. Second, a graph structure data is established and maintained based on the event stream data, defining products, components, process stations, and identification carriers as nodes, and their relationships as relational edges. Then, a graph traversal query is performed based on the graph structure data to parse the target identification carrier information and verify the integrity of the data chain before application. When the verification is successful, labeling control instructions are generated and executed to drive the actuator to complete the transfer, spatial transposition, and application of the identification carrier. After application, the application quality is confirmed through visual inspection, and the binding relationship between the product and the identification carrier is updated to the graph structure data as a new relational edge. Finally, based on the continuously updated graph structure data, forward prediction and reverse full-link traceability of data flow are achieved for online products, completing closed-loop control.

[0126] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An automated transfer and transposition method for assembling electronic products, characterized in that, Includes the following steps: S1: Collect and transform production events from multiple processes on the electronic product assembly line to generate continuous production event stream data; S2: Based on production event flow data, establish and maintain a graph structure data, in which products, components, process stations and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges; S3: Perform graph traversal queries based on graph structure data to parse out the target identifier carrier information that the current product needs to be bound to, and verify the integrity of the data chain related to the current product before the attachment operation is performed, specifically including: Obtain all established relationship edges of the current product node in the graph structure data to form a complete set of associated paths from the product's initial process to the current labeling station; The associated path set is compared with the predefined process route template to verify whether all key process station nodes are included in the associated path and whether the connection order of the relationship edges between each process station node conforms to the specification requirements of the process route template. Check the status attributes of each process station node in the associated path set, confirm that the status attributes of all necessary process station nodes are marked as qualified, and identify process station nodes with missing status or abnormal markings. The data chain integrity verification is deemed successful only if the set of associated paths completely covers the process route template and all key node statuses are qualified. Based on the binding relationship edge between component nodes and identification carrier nodes, the final target identification carrier information set is parsed and generated. The step of obtaining all established relationship edges of the current product node in the graph structure data to form a complete set of associated paths from the initial process to the current labeling station specifically includes: Starting from the current product node, perform multi-level reverse traversal along the experience relationship edge to dynamically establish a complete process path from the current labeling station node back to the starting process node; In the complete process path, based on the event timestamp attribute recorded by each process station node, the temporal continuity of each relation edge is verified to ensure that the order in which the product flows through each process station meets the predetermined production cycle requirements. Synchronously traverse the containment edges in the graph structure data, extract all component nodes associated with the current product node, and establish a mapping relationship between component nodes and corresponding process path nodes; The complete process path and component mapping relationship that have been verified by time sequence are integrated to generate a multi-dimensional set of related paths containing time dimension, process dimension and component dimension, which serves as the input basis for data chain integrity verification; S4: Only when the data link integrity verification is passed, generate and execute labeling control instructions to drive the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier; S5: After confirming that the target identification carrier has been correctly attached, submit and update the graph structure data as a new relationship edge representing the successful binding relationship between the product and the identification carrier; S6: Based on continuously updated graph structure data, it enables forward prediction and reverse full-link traceability of data flow for online products.

2. The automatic transfer and transposition method for assembling electronic products according to claim 1, characterized in that, The specific process of S2 is as follows: Real-time analysis of continuously incoming production event stream data is performed to extract event subject identifiers, event types, and event association attributes; Based on the event type, match the predefined graph operation rules and execute the corresponding node and relationship edge operations: when the event type is process completion, establish an experience relationship edge between the product node and the process station node; when the event type is configuration lock, establish an inclusion relationship edge between the product node and the component node. In response to the trigger signal of the labeling station, starting from the current product node, a bidirectional traversal is performed along the experience relationship edge and the containment relationship edge to verify the integrity of the path to the current node, and extract all the identification carrier node information associated with the end component node as the target identification carrier information.

3. The automatic transfer and transposition method for assembling electronic products according to claim 2, characterized in that, The response to the trigger signal at the labeling station involves a bidirectional traversal, starting from the current product node and proceeding along the edges of experiential and containment relationships. Specifically, this includes: Define and obtain the reverse tracing depth centered on the current product node along the experience relationship edge, dynamically set the time window based on the product process specification, and filter all process station nodes created within the time window that are directly related to the current product node; Verify that each process station node has a status attribute that represents the process being qualified, and confirm that all necessary process station nodes exist in the established associated paths, thereby verifying the integrity of the path to the current node. If and only if the path integrity verification passes, start from the current product node, traverse along the containment relationship edge to all end component nodes, and aggregate all the identifier carrier node information connected by the configuration relationship edge on each end component node to generate the final target identifier carrier information set.

4. The automatic transfer and transposition method for assembling electronic products according to claim 1, characterized in that, The specific process of S4 is as follows: Based on the target identification carrier information set generated after the data chain integrity verification is passed, the physical attribute parameters of each target identification carrier are analyzed, including the identification carrier size, material type and coordinate position in the feeder. A set of multi-dimensional control parameters, including transfer trajectory, transposition angle and attachment pressure, is dynamically generated based on physical property parameters. Based on a set of multi-dimensional control parameters, segmented motion commands are generated to drive the actuator to sequentially complete the following tasks: accurately picking up the label carrier from the feeder, transposing and adjusting it in three-dimensional space according to a specified trajectory and angle, and accurately attaching the label carrier to the designated area of ​​the product with optimized attachment pressure. During the continuous contact between the label carrier and the product surface, the matching degree between the adhesion pressure curve and the preset ideal pressure curve is analyzed in real time, and the end posture of the actuator is adjusted.

5. The automatic transfer and transposition method for assembling electronic products according to claim 4, characterized in that, The precise application of the label carrier to the designated area of ​​the product specifically includes: Based on the physical properties of the label carrier and the target attachment position, plan the continuous motion trajectory sequence of the end effector of the actuator from the feeder to the product surface. The continuous motion trajectory sequence includes the pickup preparation section, the label carrier pickup section, the spatial transposition section, and the attachment approach section. In the tag carrier pickup section, the pickup posture of the end effector is dynamically adjusted based on the peeling characteristics of the tag carrier material; In the spatial transposition segment, based on the current actual spatial posture of the product and the spatial geometric relationship between the target attachment surface of the marker carrier, the Euler angle transformation sequence of the end effector in three-dimensional space is calculated in real time to achieve a smooth transposition of the marker carrier posture. In the approach segment, based on the microscopic morphological features of the product surface fed back by the visual positioning system, the approach path and application pressure curve of the end effector are optimized to achieve progressive contact and complete adhesion between the label carrier and the product surface.

6. The automatic transfer and transposition method for assembling electronic products according to claim 1, characterized in that, The specific process of S5 is as follows: Collect multi-dimensional image data of the attached label carriers, and comprehensively analyze the positional accuracy, angle deviation and adhesion integrity of the label carriers based on the preset label carrier adhesion quality evaluation rules to generate adhesion verification results; When the attachment verification result meets the preset standard, extract the unique identifier of the current product node and the serial number information of the target identifier carrier node, and construct a binding data packet containing timestamp and workstation information; The binding relationship data packet is submitted to the graph structure database as an atomic operation transaction. A binding relationship edge with complete attributes is established between the product node and the identifier carrier node, and the status attribute of the product node is updated to "labeled" synchronously. Based on the newly established binding relationship edge-to-graph structure data, a consistency check is performed to verify the uniqueness and consistency of the correspondence between product nodes and identifier carrier nodes throughout the entire production path.

7. The automatic transfer and transposition method for assembling electronic products according to claim 1, characterized in that, The specific process of S6 is as follows: Real-time monitoring of newly added binding relationship edges in the graph structure data; starting from the product node corresponding to the newly added binding relationship edge, traversing the subsequent process path in the forward direction along the relationship edge; preloading the configuration parameters required by the downstream process and performing resource availability verification in advance. When an abnormal change in the status attribute of any process station node is detected, the corresponding abnormal node is used as the base point, and all related product nodes are traversed in reverse along the experience relationship edge to construct a complete traceability chain for the affected product. Based on the temporal attributes and topological relationships of each node in the complete traceability chain, a disposal plan including batch isolation suggestions and reprocessing paths is generated, and the disposal status of the affected product nodes is marked in the graph structure data. By continuously analyzing the frequency and distribution patterns of various relationship edges in graph structure data, bottleneck process paths in the production process can be identified.

8. An automated transfer and transposition system for assembling electronic products, characterized in that, An automated transfer and transposition method for assembling an electronic product according to any one of claims 1-7 includes: The production event stream acquisition and conversion module is used to collect and convert production events from multiple processes on the electronic product assembly line to generate continuous production event stream data. The graph structure data construction and maintenance module establishes and maintains a graph structure data based on production event flow data. In this module, products, components, process stations, and identification carriers are defined as nodes, and the relationships between nodes are defined as relationship edges. The graph traversal query and verification module performs graph traversal queries based on graph structure data, parses out the target identifier carrier information that the current product needs to be bound to, and verifies the completeness of the data chain related to the current product before the attachment operation is performed. Specifically, this includes: Obtain all established relationship edges of the current product node in the graph structure data to form a complete set of associated paths from the product's initial process to the current labeling station; The associated path set is compared with the predefined process route template to verify whether all key process station nodes are included in the associated path and whether the connection order of the relationship edges between each process station node conforms to the specification requirements of the process route template. Check the status attributes of each process station node in the associated path set, confirm that the status attributes of all necessary process station nodes are marked as qualified, and identify process station nodes with missing status or abnormal markings. The data chain integrity verification is deemed successful only if the set of associated paths completely covers the process route template and all key node statuses are qualified. Based on the binding relationship edge between component nodes and identification carrier nodes, the final target identification carrier information set is parsed and generated. The step of obtaining all established relationship edges of the current product node in the graph structure data to form a complete set of associated paths from the initial process to the current labeling station specifically includes: Starting from the current product node, perform multi-level reverse traversal along the experience relationship edge to dynamically establish a complete process path from the current labeling station node back to the starting process node; In the complete process path, based on the event timestamp attribute recorded by each process station node, the temporal continuity of each relation edge is verified to ensure that the order in which the product flows through each process station meets the predetermined production cycle requirements. Synchronously traverse the containment edges in the graph structure data, extract all component nodes associated with the current product node, and establish a mapping relationship between component nodes and corresponding process path nodes; The complete process path and component mapping relationship that have been verified by time sequence are integrated to generate a multi-dimensional set of related paths containing time dimension, process dimension and component dimension, which serves as the input basis for data chain integrity verification; The labeling control instruction generation and execution module generates and executes labeling control instructions only when the data link integrity verification is passed, driving the automated actuator to complete the transfer, spatial transposition and affixing operations of the target label carrier; The binding relationship update module, after confirming that the target identification carrier has been correctly attached, submits and updates the binding relationship that represents the successful binding between the product and the identification carrier as a new relationship edge in the graph structure data. The data flow analysis and traceability module, based on continuously updated graph structure data, enables forward prediction and reverse full-link traceability of data flow for online products.

Images