Compressor part flow prevention and error correction method and system based on identification chain verification

Abstract

This application relates to the field of error prevention technology in parts transfer, and discloses a method and system for error prevention in compressor parts transfer based on identifier chain verification. The method includes: reading compressor part identifiers and transfer fixture identifiers to form a shared record and role grid; verifying the identifier chain and exclusively reserving a primary transfer credential; verifying the compressor part identifier, fixture identifier, and machining program identifier at the near end of the equipment, and associating the control session number with the primary equipment permission; binding the equipment start event, equipment completion event, and in-machine detection event to the control session number to form a process execution witness; writing the process confirmation identifier segment, canceling the primary transfer credential, and updating the identifier chain tail segment; and reading back the control equipment exit lock status based on three types of receipts and the exit status. The technical solution of this application is used for compressor parts transfer across workstations. By using identifier chain verification and exit interlocking to maintain the correspondence between the object and the transfer record, it can enhance the certainty of anomaly blocking and the reliability of the exit release closed loop.

Description

Technical Field

[0001] This application relates to the field of error prevention technology in parts transfer, specifically to a method and system for error prevention in compressor parts transfer based on identifier chain verification. Background Technology

[0002] In existing technologies, error prevention for compressor parts transfer across workstations typically relies on machine-readable identifiers working in conjunction with equipment controllers. This involves reading information about parts, carriers, fixtures, and machining programs at loading stations, workstation entrances, and near processing equipment, and implementing release controls based on the process route and equipment feedback to meet the requirements for identity verification, process connection, and error prevention during continuous compressor parts processing. However, existing error prevention methods for compressor parts transfer have some significant shortcomings in maintaining a continuous correspondence between transfer records and on-site objects and equipment execution results.

[0003] In actual production, a single read result usually only reflects the identification combination at the time of acquisition. After the part moves with the reusable carrier and is transferred to the fixture near the equipment, factors such as cross-reading of adjacent areas, carrier occupancy fluctuations, object replacement after entry verification, and resubmission of historical release records can cause the identity relationship saved by the system to deviate from the current carrier relationship. If events such as equipment start-up, processing completion, and in-machine inspection are recorded separately, the correlation between the event's object, processing cycle, and versions of previous and subsequent processes can easily differ. The database status may continue to advance, resulting in incorrect part combinations obtaining processing permission, processes with incomplete execution evidence entering subsequent processes, or parts leaving the equipment exit before the record submission is closed.

[0004] Therefore, it is evident that existing technologies often suffer from problems such as weak stability in the cross-station flow of compressor parts and low reliability of the closed-loop process confirmation and export release. These are the shortcomings of existing technologies.

[0005] In view of this, it is necessary to provide a method and system for preventing errors in the flow of compressor parts based on identifier chain verification, so as to solve the above-mentioned defects in the prior art. Summary of the Invention

[0006] The purpose of this application is to address the shortcomings of the prior art, namely, the weak stability of compressor parts flow status across workstations and the low reliability of the closed loop between process confirmation and export release, by providing a method and system for preventing errors in the flow of compressor parts based on identifier chain verification, so as to solve the above-mentioned technical problems.

[0007] To achieve the above objectives, this application provides the following technical solution: Firstly, this application provides a method for error prevention in the flow of compressor parts based on identifier chain verification, including: Step S1: Read the compressor part identifier and the transfer tool identifier at the loading position and the work station entrance, form a co-occupancy record according to the continuous reading interval and the tool occupancy status, and write it into the role grid according to the reading role and role boundary version; Step S2: Verify the identifier chain based on the role grid, the tail segment of the identifier chain, and the process route version to form a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserve one flow credential. Step S3: Read the compressor part identifier, fixture identifier, and machining program identifier at the near end of the equipment, match the reading results with the successor candidate, generate a control session number, and associate it with a primary equipment license and a primary transfer credential; Step S4: Start processing with a single equipment license, and bind the equipment start event, equipment completion event, and in-machine detection event to the control session number, processing cycle identifier, and credential sequence number to form a process execution witness; Step S5: Read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. Step S6: Control the device's exit lock status according to the three types of receipts. When all three types of receipts are complete, the receipt content is consistent, and the version of the updated receipt record before the update is consistent with the version submitted as the baseline, release the device's exit lock status based on the exit status readback. If any condition is not met, maintain the device's exit lock status and freeze the transfer voucher once.

[0008] By adopting the above technical solution, the object identity, positional relationship, process continuity, processing execution, and export release in the cross-workstation flow of compressor parts are incorporated into the same identification chain verification closed loop. This achieves continuous constraints on digital records, on-site bearing relationships, equipment control status, and process submission results. It can maintain consistency in the flow judgment basis even when the flow transfer tool is reused, processing resources are switched, and the workstation status changes continuously. This meets the requirements of strong stability of the cross-workstation flow status of compressor parts and high reliability of the closed loop of process confirmation and export release.

[0009] Specifically, by defining the entry relationship with bounded shared records, instantaneous read results are transformed into continuous carrying basis, reducing state interference caused by serial reading, idle passage, and object replacement, and ensuring consistent judgment across different read positions; by constraining exclusive continuation with the identifier chain tail segment and process route version, the process status continues only to the matching expected read role, preventing out-of-order, duplicate occupation, and credential replay before processing starts; by binding the control session number with equipment near-end verification, it is beneficial to fix the processing object and resource combination, compress the risk window of replacement after entry verification, and enhance the stability of the correspondence between primary equipment permission and on-site status; and by using the same control session... The system organizes equipment startup events, equipment completion events, and in-machine testing events, prompting decentralized feedback to form clearly attributed and sequentially sequential process execution witnesses, avoiding a single signal pushing the status forward. Versioned batch submissions generate three types of receipts, facilitating verification of the consistency and version of the objects written to the process confirmation identifier segment, the cancellation of the primary transfer voucher, and the update of the identifier chain tail segment, improving the completeness of status submission. Furthermore, interlocking is implemented through receipt consistency and exit status rereading, keeping the equipment exit status locked and freezing the primary transfer voucher under abnormal conditions, ensuring that abnormal handling and exit release are executed according to clear boundaries, forming a closed loop of error prevention from the entrance to the exit.

[0010] Preferably, step S2 specifically includes: The co-occupancy records in the role grid are arranged according to the read role and the role boundary version to form the role sequence corresponding to the same compressor part identifier, and then compared with the end role of the record at the end of the identifier chain. Verify the validity of the role boundary version and the process route version, obtain the next role of the end role according to the valid process route version, match the next role with the first valid role in the role sequence, and filter out co-occupancy records with invalid versions or non-adjacent roles from the role sequence. Compare the compressor part identifier and transfer device identifier in the retained record with the tail segment of the identifier chain, and find the overlapping interval of the continuous reading interval; Based on the comparison results of the two identifiers, the vehicle occupancy status, and the overlapping area, a succession candidate containing the expected reading role and credential succession sequence number is formed; Query the reserved workstations for a single transfer voucher, bind the single transfer voucher that has not been registered for a reserved workstation to the successor candidate, and write the single transfer voucher that has been registered for other workstations into the cross-workstation conflict record.

[0011] Based on the above scheme, joint constraints are implemented on the validity of role succession and the exclusive use of one-time transfer vouchers. This ensures that the on-site carrying relationship only enters adjacent workstations along the valid process route, and retains the conflict boundary when competing across workstations. This can prevent invalid versions, non-adjacent roles, or insufficiently overlapping records from participating in the succession, suppress the parallel reservation of the same one-time transfer voucher, and improve the consistency of identification link succession judgment and the clarity of anomaly tracing.

[0012] Preferably, step S3 specifically includes: According to the continuous reading interval of the near end of the equipment, the compressor part identifier, fixture identifier and machining program identifier are grouped into the same near end window, and the near end window is closed according to the clamping in place signal; The system continuously matches the read results appearing in the near window with the succession candidates one by one, and verifies the succession sequence number of the voucher. When the compressor part identifier is replaced, the reservation of a transfer voucher is cancelled. When the fixture identifier or machining program identifier does not match, the transfer voucher is frozen. After the reading result matches the continuation candidate, the first-pass credential is changed from the reserved state to the occupied state, a control session number is generated, and the continuation candidate, the near window, and the credential continuation sequence number are associated with the control session number. A device license is generated based on the control session number and written to the device controller. When the same transfer credential corresponds to another control session number or the read result changes, the unexecuted device license is revoked, and the occupancy status of the transfer credential is maintained.

[0013] In the above scheme, the combination of processing objects is fixed by the near-end window of the equipment, and the verification boundary is limited by the clamping position signal. The mismatch between object replacement and processing resources will enter the reserved cancellation or the freezing branch of the first-time transfer certificate respectively, so as to prevent the abnormal state from continuing to be converted into the start permission. The control session number is used to maintain the unique association between the first-time equipment license and the object combination. This can reduce the risk window of part replacement, fixture change or program change in the entry verification, prevent parallel sessions and residual licenses from triggering false start, and improve the determinism of the equipment start boundary and the reliability of the correspondence between processing objects.

[0014] Preferably, step S4 specifically includes: Receive the processing cycle identifier returned by the device controller and associate the processing cycle identifier with the control session number and credential sequence number; Obtain equipment start-up and equipment completion events from the equipment controller, obtain in-machine detection events from the in-machine detection interface, and write the three types of events into the execution window corresponding to the processing cycle identifier; Verify the control session number, processing cycle identifier, and credential continuation number of the three types of event records. Form an event sequence for the three types of events that are consistent in all three aspects according to the collection order, and write any event that is inconsistent in any aspect into the conflict event set. Based on the successive verification results of the event sequence, a process execution witness is formed. When an event is missing or the successive relationship is not established, the corresponding execution window is closed, and the occupancy status of one transfer credential is retained.

[0015] Based on the above processing, three types of events are collected around the same processing cycle and the event attribution is verified, so that the current processing process forms a process execution witness with a unified source and continuous sequence. This can isolate feedback with inconsistent attribution, promptly expose missing events or broken continuity, avoid a single completion signal driving the process status to update prematurely, and maintain the occupation of one flow credential when the evidence is incomplete, thereby enhancing the verifiability of execution results and the reliability of state transition.

[0016] Preferably, the step of writing the three types of events into the execution window corresponding to the processing loop identifier specifically includes: The execution window is opened based on the equipment start event and stopped based on the equipment completion event. The execution window receives new on-machine detection events and associates the start and end times of the execution window with the processing cycle identifier. The on-machine detection events in the execution window are arranged according to the collection time. The control session number and credential sequence number of the on-machine detection event record are compared. On-machine detection events that do not match are written into the conflict event set. Based on the acquisition time of the equipment completion event, identify the adjacent detection confirmation event from the in-machine detection events that have never been written into the conflict event set, and compare the processing program identifier of the detection confirmation event with the processing program identifier of the primary equipment permission record; When the processing program identifiers of the detection and confirmation events are consistent, the equipment start event, equipment completion event, and detection and confirmation event are identified as the valid event set of the execution window corresponding to the processing cycle identifier. When the detection and confirmation event is missing or the processing program identifiers are inconsistent, the execution window is marked as an event gap.

[0017] Based on the above processing logic, the on-machine detection event is constrained by the execution window, and the detection confirmation event is determined according to the equipment completion time. This makes the detection results close to the process termination point the confirmation basis for the current processing cycle. It can reduce misjudgments caused by the mixing of inconsistent program information across cycles and the incorporation of inconsistent program information. It can also form a clear event gap when the detection evidence is missing or mismatched, and improve the time consistency of process execution witnessing and the verifiability of detection conclusions.

[0018] Preferably, step S5 specifically includes: Based on the process execution witness, a submission batch number and a submission sequence are generated. The write sequence number, reversal sequence number and update sequence number are obtained from the submission sequence. The chain tail version that identifies the chain tail segment is recorded as the submission base version. The control session number and processing cycle identifier are assigned to the session item, the credential continuation sequence number and continuation candidate are assigned to the continuation item, and the session item and continuation item are combined to form the submission payload. Write the batch number into the process confirmation segment and obtain a write receipt that records the batch number, write sequence number, and submitted payload. Once a transfer voucher is cancelled based on the submitted batch number, a cancellation receipt is obtained after the voucher sequence number matches, recording the submitted batch number, cancellation sequence number, and submitted load. After the tail version is consistent with the submission baseline version, update the tail segment to the process confirmation segment and obtain the update receipts for the submission batch number, update sequence number, submission payload, tail version before update and tail version after update. Verify the write receipt, cancellation receipt, and update receipt, and generate a submission status based on the verification results.

[0019] By adopting the above technical solution, the three status changes are uniformly constrained by the batch number and the base version of the submission, so that the operation is completed within the same object scope, session relationship and sequence number boundary. This can avoid the partial write success being mistaken for the overall submission completion, identify the changes at the end of the concurrent chain and the shift in the order of operations, keep the process execution results, the status of the first-time transfer voucher and the tail segment of the identification chain progressing synchronously, and provide a traceable receipt basis for the submission status determination.

[0020] As a preferred option, the steps for verifying the write receipt, the cancellation receipt, and the update receipt specifically include: The three types of receipts are grouped into a receipt set. Receipts that are not from the current submission batch number are filtered out by receipt type and corresponding serial number; Compare the session items and continuation items in the receipt set, and record any inconsistency as a load conflict; Compare the order of writing sequence number, canceling sequence number, and updating sequence number, and record duplicate or reversed sequence numbers as operation sequence number conflicts; Compare the sequential number of the write-off receipt and the document for witnessing the execution of the process. Record the smaller former as a historical replay and the larger former as a sequential conflict. Compare the pre-update tail version of the update receipt with the submission baseline version. Record inconsistencies in the pre-update tail version, operation sequence number conflicts, or historical replays as timing conflicts, and record continuation conflicts as payload conflicts. When there are no receipt gaps, load conflicts, or timing conflicts in the receipt set, the receipt set is recorded as a complete submission status; When any issue exists, the receipt set is marked as an incomplete submission status, and the corresponding issue is written to the submission status.

[0021] Based on the above scheme, cross-checking of batch attribution, load relationship and operation sequence is carried out on the three types of receipts. This prevents receipts from different sources but with similar surface fields from being spliced ​​into a complete submission. Receipt gaps, historical replays and continuation offsets are transformed into distinguishable abnormal evidence. This can promptly identify the reuse of old receipts, reverse operations and concurrent state mixing, and prevent inconsistent records from driving the identification chain to continue updating. This enhances the replay resistance, concurrency identification capability and subsequent control reliability of the submission status determination.

[0022] Preferably, step S6 specifically includes: Associate the equipment's export lock status with the submission batch number to generate an export lock serial number, and write the complete or incomplete submission status into the export control record. Based on the incomplete submission status, the equipment exit lock status is maintained and the transfer voucher is frozen once. Subsequent equipment licenses corresponding to the same compressor part identifier are refused, and the exit lock serial number is written into the rejection record. Based on the complete submission status, compare the process confirmation identifier segment of the updated receipt record with the process confirmation identifier segment of the identifier chain tail record. If they match, generate an unlocking instruction containing the exit lock sequence number and control session number. Write the unlock command to the device controller, receive the exit status readback containing the exit lock sequence number and control session number, compare the two items in the exit status readback with the unlock command, when the two items match, write the unlock confirmation and release the device exit lock status, when either item does not match, write the lock confirmation and keep the device exit lock status.

[0023] In the above scheme, a direct control relationship is established between the submission status and the equipment exit lock status. When the submission is incomplete, the physical exit remains restricted and subsequent equipment permission is rejected. A complete submission still requires a two-way correspondence between the unlock command and the exit status readback to form an unlock confirmation. This can avoid the situation where a one-sided database update or the command is sent is regarded as a release completion. It can also promptly identify the deviation between the control session number and the actual response of the equipment, ensuring that the compressor parts only leave the workstation when the record is closed and the equipment response is consistent, thereby improving the certainty of abnormal blocking and the reliability of exit release feedback.

[0024] Preferably, the steps of maintaining the device's exit lock state and freezing the transfer credentials once based on the incomplete submission status specifically include: Based on the submitted batch number, obtain the receipt set and process execution witness, and compare the control session number, processing cycle identifier, and document continuation number in the receipt set with the corresponding content of the process execution witness; Obtain the processing cycle identifier and voucher sequence number from the equipment controller, compare the two items with the corresponding items in the process execution witness, and record any inconsistency as an execution conflict; When all items are consistent and only the receipt is missing, the corresponding problem is recorded as a recoverable gap. The equipment exit is kept locked and the transfer voucher is frozen once. The missing operation is performed using the submitted batch number, and the new receipt is written into the receipt set. When load conflicts, timing conflicts, or execution conflicts exist, the corresponding issues will be recorded as isolation conflicts, the missing operations will be refused to be executed, the equipment exit will be kept locked and the transfer voucher will be frozen, and the compressor part identification and submission batch number will be written into the isolation record.

[0025] In the face of incomplete submission scenarios, the consistency between the verification receipt and the actual execution of the equipment is verified by the above processing logic. In cases where there is only a gap in the receipt and the execution relationship has not changed, the submission batch number is used for recovery. However, load, timing or execution relationship offsets are isolated and handled. This can avoid unnecessary stagnation caused by recoverable gaps and prevent the spread of substantial conflicts through supplementary execution, thus taking into account production continuity, the accuracy of abnormal handling and the safety boundary of parts flow.

[0026] Secondly, this application also provides a compressor parts transfer error prevention system based on identifier chain verification, comprising: The co-occupancy write module is used to read compressor part identifiers and transfer fixture identifiers at loading positions and workstation entrances, form co-occupancy records according to continuous reading intervals and fixture occupancy status, and write them to the role grid according to the reading role and role boundary version. The identifier verification module is used to verify the identifier chain based on the role grid, the tail of the identifier chain, and the process route version, forming a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserving one flow credential. The session authorization module is used to read compressor part identifiers, fixture identifiers, and machining program identifiers at the near end of the equipment, match the reading results with success candidates, generate a control session number, and associate a device authorization with a transfer credential. The witness generation module is used to start processing with a single equipment license, and bind the equipment start event, equipment completion event and in-machine detection event to the control session number, processing cycle identifier and credential sequence number to form a process execution witness; The confirmation and submission module is used to read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. The export control module is used to control the export lock status of the equipment according to three types of receipts. When all three types of receipts are complete, the content of the receipts is consistent, and the version of the chain tail before the update of the updated receipt record is consistent with the version submitted as the baseline, the export lock status of the equipment is released based on the export status readback. When any condition is not met, the export lock status of the equipment is maintained and the transfer voucher is frozen once.

[0027] Based on the above technical solution, this application has at least the following technical effects: This application provides a method and system for preventing errors in the flow of compressor parts based on identifier chain verification. By incorporating the object identity, positional relationship, process continuity, processing execution, and export release in the cross-station flow of compressor parts into the same identifier chain verification closed loop, continuous constraints are achieved on digital records, on-site bearing relationships, equipment control status, and process submission results. This ensures consistency in the flow judgment criteria even with reuse of flow transfer tools, switching of processing resources, and continuous changes in station status. It meets the requirements of strong stability of the cross-station flow status of compressor parts and high reliability of the closed loop for process confirmation and export release. Attached Figure Description

[0028] To facilitate the explanation of the technical solutions of the embodiments of this application, the accompanying drawings used in this application will be briefly described below.

[0029] Figure 1 This is a flowchart of a compressor parts transfer error prevention method based on identifier chain verification provided in this application; Figure 2 This is a schematic diagram of a compressor parts transfer error prevention system based on identifier chain verification provided in this application.

[0030] The modules are: 1. Shared write module, 2. Identifier verification module, 3. Session permission module, 4. Witness formation module, 5. Confirmation and submission module, and 6. Export control module. Detailed Implementation

[0031] The embodiments of this application will now be described with reference to the accompanying drawings. It should be understood that the following embodiments are only used to illustrate the technical solutions of this application and do not limit the scope of protection of this application. Those skilled in the art can make equivalent substitutions or improvements based on the disclosure of this application.

[0032] It should be noted that in the description of this application, the terms "comprising," "including," "having," and their synonyms are used to indicate the presence of the described features, structures, steps, operations, elements, components, or combinations thereof, but do not exclude the presence of other features, structures, steps, operations, elements, components, or combinations thereof.

[0033] It should be noted in advance that, for ease of understanding of the technical solution of this application, the following explanations are provided regarding the terminology and related technologies involved in this application: 1. Process route: In discrete manufacturing, it is a technical document or data object used to describe the processes, process sequence, work centers, and corresponding production resources and tools that need to be performed when producing a certain type of part. It usually serves as the basis for production task arrangement, process succession verification, and manufacturing process execution.

[0034] 2. In-machine inspection: refers to measurement or inspection performed while the workpiece is still in the processing equipment or in the current clamping state. Information on the workpiece position, size, shape or processing result is obtained through in-machine probes, sensors or detection interfaces. This is used to reduce the reference changes introduced by reclamping and to provide on-site data for confirming the status of the current processing cycle.

[0035] 3. Equipment Controller: An industrial control unit used to receive field input signals, process operating conditions according to preset control logic, and issue start / stop, interlock, or state switching commands to the actuators. Common implementation forms include PLC (Programmable Logic Controller) and CNC equipment controller.

[0036] To address the issues of unstable connection between recorded status and on-site objects, equipment execution, and physical release, which makes it difficult to meet the actual needs of continuous processing sites for consistent flow basis and reliable anomaly blocking, resulting in weak stability of compressor parts flow status across workstations and low reliability of closed-loop process confirmation and exit release, this application discloses a compressor parts flow error prevention method and system based on identifier chain verification. By introducing a collaborative verification mechanism of object relationship, succession permission, execution evidence, and submission receipt, the processing permit is kept consistent with the current processing object combination, and the process status is only continued under complete witness and consistent receipt, thereby improving the verifiability of compressor parts flow status across workstations, the determinacy of blocking abnormal branches, and the closed-loop reliability of exit release.

[0037] like Figure 1 As shown, this embodiment provides a method for error prevention in the flow of compressor parts based on identifier chain verification, including: Step S1: Read the compressor part identifier and the transfer tool identifier at the loading position and the work station entrance, form a co-occupancy record according to the continuous reading interval and the tool occupancy status, and write it into the role grid according to the reading role and role boundary version; Step S2: Verify the identifier chain based on the role grid, the tail segment of the identifier chain, and the process route version to form a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserve one flow credential. Step S3: Read the compressor part identifier, fixture identifier, and machining program identifier at the near end of the equipment, match the reading results with the successor candidate, generate a control session number, and associate it with a primary equipment license and a primary transfer credential; Step S4: Start processing with a single equipment license, and bind the equipment start event, equipment completion event, and in-machine detection event to the control session number, processing cycle identifier, and credential sequence number to form a process execution witness; Step S5: Read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. Step S6: Control the device's exit lock status according to the three types of receipts. When all three types of receipts are complete, the receipt content is consistent, and the version of the updated receipt record before the update is consistent with the version submitted as the baseline, release the device's exit lock status based on the exit status readback. If any condition is not met, maintain the device's exit lock status and freeze the transfer voucher once.

[0038] This embodiment transforms the relationship between compressor parts and conveying equipment from a single read result into a sustainable verification basis through co-occupancy records with continuity and positional boundaries. This suppresses relationship drift caused by cross-reading, no-load entry, and object replacement. Exclusive continuation is formed using the tail segment of the identifier chain and the process route version, ensuring that confirmation results continue only along the matched expected read role. This helps stabilize process boundaries and prevents interference from out-of-order continuation, parallel occupation, and credential replay on the processing sequence. Binding control session numbers to near-end equipment verification ensures that primary equipment permissions correspond to processing objects and resource combinations, reducing the risk window of replacement after entry verification and enhancing the determinism of equipment startup conditions. Organizing equipment startup events, equipment completion events, and in-machine detection events around the same control session number ensures that dispersed equipment feedback forms a clear attribution and sequential continuity. Witnessing process execution avoids single signals driving status forward and provides verifiable on-site execution evidence for process confirmation. Submitting a baseline version and batch number yields three types of receipts, ensuring that the writing of the process confirmation identifier, the cancellation of a single transfer document, and the update of the identifier chain tail segment are all verified against the same object and version. This helps identify incomplete submissions and concurrent changes, improving the consistency of submission results. Based on the three types of receipts, version consistency, and exit status readback, exit interlocking is implemented to maintain the equipment exit locked state and freeze the single transfer document under abnormal conditions, preventing compressor parts from continuing to circulate when the digital status is not closed. Overall, through the continuous coordination of the aforementioned technical steps, a closed-loop constraint is formed from entry relationship confirmation to equipment exit release, improving the stability of compressor parts' cross-station transfer status, the certainty of abnormal blocking, and the adaptability to on-site operation.

[0039] The above steps will be specifically described below based on the embodiments of this application.

[0040] In step S1, the core task is to establish a continuous co-occurrence relationship between the compressor part identifier and the transfer fixture identifier from the loading position and the workstation entrance, and to solidify the reading position, fixture occupancy status, and role boundary version into a role grid that can be verified and read by subsequent identifier chains. The reading channel is triggered by the workstation controller according to a unified acquisition clock. The loading position covers the stage when the compressor part is loaded into the transfer fixture, and the workstation entrance covers the stage when the transfer fixture enters the controlled workstation. Both positions retain the reading channel identity, acquisition time, identifier identity, fixture occupancy status, and role boundary version.

[0041] In some embodiments of this application, compressor part identifiers are set on machine-readable recording carriers corresponding to pistons, swashplates, planetary disks, valve cores, or moving and stationary scroll disks, while transfer device identifiers are set on pallets, turnover boxes, positioning frames, or automatic loading and unloading carriers. The carrier occupancy status is obtained by photoelectric switches, proximity switches, weight detection components, or load confirmation signals, and is written using the same acquisition clock as the reading event. After the reading channel begins acquisition, compressor part identifiers and transfer device identifiers are read at the loading position and workstation entrance. The reader continuously retains the times of the first stable appearance, continuous presence, and departure from the reading area for each identifier; single, occasional readings do not directly enter the shared occupancy record.

[0042] Specifically, when the gap between adjacent acquisition cycles for the same identifier does not exceed twice the stable acquisition cycle of the reading channel, the station controller merges the records on both sides of the gap to form a continuous reading interval; when the gap exceeds this boundary, the station controller closes the previous continuous reading interval and establishes a new continuous reading interval. When multiple compressor part identifiers or multiple transfer device identifiers appear in the same acquisition phase of the same reading channel, each identifier retains its own continuous reading interval, and the acquisition phase is written into the conflict state to prevent adjacent device reads from being mistakenly merged into the target identifier.

[0043] Based on this, a co-occupancy record is formed according to the continuous read intervals and the vehicle occupancy status. The formation process of the co-occupancy record simultaneously reads the overlap length, individual duration, effective vehicle occupancy amount, and maximum time deviation between the start and end boundaries of the two continuous read intervals. This ensures that records with short actual co-occurrence times, invalid vehicle occupancy, or abnormal entry / exit order do not achieve high co-occupancy confidence. For example, the co-occupancy confidence of the current continuous read window can be written as: , in, Indicates the first The co-occupancy confidence level of compressor part identifiers and transfer device identifiers within a continuous reading window. This indicates the continuous reading range of compressor part identification. This indicates the continuous reading range identified by the stream carrier. Indicates the effective amount of space occupied by the vehicle. This indicates the maximum time deviation between the start and end boundaries of the two identifiers. This represents the attenuation scale of the shared occupancy confidence level on the deviation of the identification start and end boundaries. Based on this, once the shared occupancy confidence level reaches the fixed boundary corresponding to the loading position or workstation entrance, the workstation controller retains the compressor part identifier, transfer tool identifier, continuous reading interval, tool occupancy status, and shared occupancy confidence level. When the shared occupancy confidence level is lower than the corresponding fixed boundary, the workstation controller retains the original reading event and generates a supplementary reading status. The two types of fixed boundaries are determined based on the reading area length, transfer tool passage speed, position occupancy detection response delay, and role boundary calibration results, and are written into the role boundary version, remaining unchanged during online operation regardless of a single result.

[0044] In some embodiments of this application, after the loading position completes the co-occupancy record, a loading read role is assigned; after the workstation entry completes the co-occupancy record, a workstation entry read role is assigned. The co-occupancy record is written to the role grid according to the read role and role boundary version. The writing process first searches for a fixed role in the role boundary version based on the read channel identity. Then, the co-occupancy confidence quantity, vehicle occupancy validity quantity, channel role consistency quantity, role boundary version validity quantity, and missing or conflict status code are written to the same role grid record in a fixed field order. A continuous read window reference and a role boundary version reference are written at the end of the record, enabling the source of each status to be read back when the role sequence is arranged.

[0045] It should be noted that when a character boundary version changes, existing continuous reading windows will continue to use the character boundary version corresponding to when the window was opened, while subsequent new windows will use the updated character boundary version. After a reader is replaced, antenna position is adjusted, or the visual reading field of view changes, the new character boundary version will only take effect after the fixed compressor part identifier and fixed transfer vehicle identifier have entered each reading character in sequence and completed the serial reading boundary scan. Reading events where a unified time reference cannot be obtained, vehicle occupancy status is missing, or character boundary version is invalid are retained in the supplementary certification record and are not written into a valid character sequence that can be used for continued identification linking.

[0046] Thus far, step S1 has completed the step-by-step transformation of compressor part identifiers and transfer vehicle identifiers from the original reading event to continuous reading intervals, shared records, and role grid records. It also defines the physical location and effective boundaries of each record by vehicle occupancy status, reading channel space role, and role boundary version, providing a locatable input basis for step S2 to verify the identifier chain according to the role sequence and form succession candidates.

[0047] In step S2, the core task is to convert the physical co-occupancy record in the role grid, along with the identifier chain tail segment, process route version, and primary transfer document status, into a unique and usable successor candidate, and to identify the specific source of conflict under conditions of concurrent reading at multiple workstations, role boundary changes, and short-term fluctuations in identifiers. This step reads the role grid record formed in step S1, the identifier chain tail segment retained after confirmation by the previous process, the process route version corresponding to the compressor part, and the reservable status of the primary transfer document.

[0048] In this embodiment, the identifier chain is verified based on the role grid, the tail segment of the identifier chain, and the process route version to form a succession candidate containing the expected reading role and the credential succession sequence number, and a single flow credential is exclusively reserved. Before the identifier chain verification begins, the workstation controller obtains the co-occupancy record with the original reading event reference from the role grid, obtains the most recently confirmed compressor part identifier, flow transfer tool identifier, end role, process route version, credential succession sequence number, and tail version from the identifier chain storage area, and obtains the expected reading role, credential succession sequence number, reserved workstation, occupancy status, and cancellation status from the credential storage area. If any object is missing, version reading fails, or credential status cannot be read back, the current record enters the supplementary credential reading state, no succession candidate is generated, and no permission basis is transmitted to the near end of the equipment.

[0049] Specifically, the co-occupancy records in the role grid are arranged according to the reading role and role boundary version, forming a role sequence corresponding to the same compressor part identifier, and then compared with the last role of the record at the end of the identifier chain. During arrangement, the records are first grouped by role boundary version, and then stably sorted according to the start time of the continuous reading window and the record writing order. When multiple records for the same compressor part identifier appear in the same reading role, the record with higher co-occupancy confidence and valid vehicle occupancy is retained as the master record, and the remaining records are retained as evidence of repeated readings. The beginning of the role sequence corresponds to the current workstation entrance, and the end of the role sequence corresponds to the last reading role that can be confirmed before the nearest point of the equipment.

[0050] Further, verify the validity of the role boundary version and the process route version. Obtain the next role for the end role based on the valid process route version. Match the next role with the first valid role in the role sequence. Remove co-occupancy records with invalid versions or non-adjacent roles from the role sequence. The validity quantity of the role boundary version is determined by whether the role boundary version used by the record is in an activation cycle. The validity quantity of the process route version is determined by whether the process route version carried by the tail segment of the identifier chain is still applicable to the current compressor part. The adjacency permission relationship between the end role and the first valid role, the validity quantities of the two types of versions, and the role sequence deviation together form the role succession compatibility quantity, which can be written as: , in, Indicates the first The number of consecutive candidate roles for compatibility. Indicates the first The number of adjacent permits for the process route of the last candidate corresponding to the first valid role. Indicates the first The effective quantity of the role boundary version adopted by each successive candidate. Indicates the first The effective quantity of each successive candidate process route version. Indicates the first The role order deviation of the successive candidates. This indicates the attenuation scale of role succession compatibility quantity to role sequence deviation. During this process, records with a role succession compatibility quantity of 0 are directly written to the version invalidation or role non-adjacent state; records with a role succession compatibility quantity between 0 and the permission boundary are retained as records awaiting recertification, requiring a new role boundary version or process route version; records with a role succession compatibility quantity reaching the permission boundary are entered into the identifier comparison. When the process route version is updated, compressor parts that have already formed the tail segment of the identifier chain continue to use the version of the tail record, while compressor parts that have not yet formed an initial bearing relationship use the new version, preventing the same compressor part from using different permission relationships between adjacent processes.

[0051] Furthermore, the compressor part identifier and the transfer vehicle identifier in the retained records are compared with the tail segment of the identifier chain to determine the overlapping interval of the continuous reading intervals. The compressor part identifier comparison uses field-by-field consistency between the identifier identity and the drawing version, while the transfer vehicle identifier comparison uses field-by-field consistency between the vehicle identity and the allowed carrying range. The overlapping interval of the continuous reading intervals is determined based on the common coverage area of ​​the two identifiers under the same role. If the overlapping interval is empty, the vehicle occupancy is invalid, or either identifier is inconsistent with the tail segment of the identifier chain, the current record will not be included in the valid candidate list.

[0052] Based on this, and according to the comparison results of the two identifiers, the vehicle occupancy status, and the overlapping area, a succession candidate is formed, containing the expected reading role and the credential continuation number. The expected reading role is obtained based on the end role recorded in the identifier chain tail segment and the valid process route version. The credential continuation number is generated according to the process confirmation order based on the previous credential continuation number. The succession candidate simultaneously retains the role grid record reference, identifier chain tail segment reference, overlapping area, vehicle occupancy status, role continuation compatibility, expected reading role, and credential continuation number, so that subsequent near-end verification by the equipment can return to each input of this candidate.

[0053] It should be further explained that while deterministic adjacency permission can filter out explicit out-of-order, version invalid, and inconsistent identifier records, when there are cross-reads in adjacent reading channels, partial missing roles in the role sequence, fluctuations in the boundaries of continuous reading intervals, and multiple candidates sharing similar identifier fragments, the fixed boundary can only provide a result of allowing continued comparison or prohibiting continued continuation, and cannot reliably distinguish between reading fluctuations requiring supplementary certification and object conflicts that need to be isolated. To allow the above-mentioned approximate states to enter different processing paths, while maintaining the process route permission, the exclusive status of the first-time transfer document, and the fact that the tail segment of the identifier chain is not rewritten by the probability result, a role continuation conflict discrimination model is set up; the model only performs auxiliary discrimination on continuation candidates that pass the basic identifier comparison, outputting the probabilities of legal continuation and various conflict states, and the final candidates still need to satisfy the deterministic permission boundary.

[0054] The model input first concatenates the role raster record vector, identifier chain tail segment summary vector, process route version code vector, first-transfer voucher reservable state quantity, and role succession compatibility quantity corresponding to the current succession candidate in a fixed field order. The model input vector of the current succession candidate input role succession conflict discrimination model can be written as: , in, Indicates the first The model input vector of the successive candidate input role succession conflict discrimination model. Indicates the first The first successor candidate corresponds to the first Each role raster sequence encoding vector is obtained by encoding the currently associated role raster record according to the read role, the start time of the continuous read window, and the record writing order. Indicates the first Each continuation candidate corresponds to a summary vector of the tail segment of the identifier chain. Indicates the first The process route version encoding vector corresponding to each successive candidate. Indicates the first Each successor candidate can reserve a state quantity for a single transfer credential.

[0055] Furthermore, a role attention layer is set at the model entry point. This layer does not change the process route version, the expected reading role, or the credential continuation number; it only assigns record contributions based on the persistence, missing state, conflict state, and coupling relationship of the role raster record with the model input vector. For example, the role record attention weight assigned by the current continuation candidate to the current role raster record can be written as: , in, Indicates the first The first successor candidate pair Each character grid record assigns a character record with a weight. This represents the trainable transpose projection vector of the role attention layer. The role attention projection matrix represents the input vector of the model. The character attention projection matrix represents the character grid record vector. Indicates the first The number of consecutive candidate associated role raster records, Indicates the first Each character's raster record vector, In the normalized summation, the first... Each role's raster record vector.

[0056] Based on this, the role attention layer aggregates all role raster records with the same consecutive candidate association according to the role record attention weight. Records that appear continuously and have a stable co-occupancy confidence value receive higher contributions, while short-term cross-reads, missing, or conflicting records are still retained but their contributions are suppressed. For example, the first... A candidate context vector is formed by aggregating role sequence information from consecutive candidates. It can be written as: , In some embodiments of this application, the candidate context vector and the model input vector enter a gated state update. The gated state update simultaneously retains the previous memory state formed by the tail segment of the identifier chain and the new state formed by the current role sequence, avoiding overwriting already confirmed tail relationships with a single read. The candidate update gate vector for the current successor candidate can be written as: , in, Indicates the first A succession of candidate update gate vectors, This represents the input transformation matrix of the candidate update gate. This represents the bias vector of the candidate update gate.

[0057] Furthermore, the candidate update gate vector controls the writing ratio of the previous memory state and the current candidate state dimension by dimension. When the role continuation compatibility, the number of states that can be reserved with a single transfer credential, or the number of valid versions decrease, the previous memory state maintains a high proportion and prevents the current abnormal record from driving a legitimate continuation; when the current role sequence is complete and the deterministic permission is established, the current candidate state enters the candidate memory state. The candidate memory state can be written as: , in, Indicates the first The candidate memory state is formed after gating and updating a successive candidate. Indicates the first The preceding memory state is formed after reading the tail segment of the successive candidate read identifier chain. This represents the transformation matrix that allows the model input vector to enter the candidate memory state. This represents the transformation matrix that allows a candidate context vector to enter a candidate memory state. The bias vector represents the candidate memory state.

[0058] Furthermore, the state output layer maps candidate memory states to the state probabilities corresponding to: valid continuation, non-adjacent roles, invalid role boundary versions, invalid process route versions, inconsistent compressor part identifiers, inconsistent transfer tool identifiers, insufficient overlapping intervals, unreservable credentials, and cross-station conflicts. The predicted probability of the current continuation candidate belonging to each continuation state can be written as: , in, Indicates the first The next candidate belongs to the first The predicted probability of each successive state. This represents the transformation matrix for a candidate memory state to enter the output state layer. This represents the bias vector of the state output layer.

[0059] In some embodiments of this application, training samples are based on event trajectories that have completed process confirmation and can trace the tail segment of the identification chain, role grid records, and the status of a single transfer credential. Valid samples require continuous role sequences, valid versions of both types, identical identifiers, overlapping intervals reaching boundaries, and credentials not reserved at other workstations. Conflict samples include adjacent vehicle cross-reading, part replacement, route out-of-order, historical credential replay, and cross-workstation concurrent reservation. Controlled fault injection changes only one object relationship each time, and trajectories lacking on-site verification conclusions are not included in the training batch.

[0060] The model employs offline, phased training. Phase 1 fixes the process route permission mask and credential exclusivity rules, training only the role attention layer, gating state update layer, and state output layer. Phase 2 incorporates read boundary fluctuations, missing role records, and cross-read samples from adjacent channels. Phase 3 adds different process route versions and low-frequency compressor part models. The state classification error and process route structure constraints together form the classification and process route structure constraint objectives, which can be written as: , in, This represents the classification and process route structure constraint objective of the role succession conflict discrimination model. This indicates the number of training samples included in a single model training batch. This indicates the number of successive state categories used in model training. Indicates the first batch of model training The training sample pair for the th training sample A true label for a continuation state. Indicates the first batch of model training The training sample belongs to the first... The predicted probability of each successive state. Indicates the first batch of model training The training sample pair for the th training sample A process route permission mask for a sequential state. This represents the training coordination coefficient of the process route structural constraint term.

[0061] Building upon this, the classification and process route structure constraints continue to be coupled with the first-passage document state constraints and the probability interval constraints between legal states and the highest conflict state. This ensures that model parameter updates are simultaneously controlled by successive state classification, prohibited state edge output, document state prediction, and state interval, which can be written as: , in, This represents the model training objective of the role succession conflict discrimination model. This represents the training coordination coefficient of the status constraint term of a single transaction credential. The training coordination coefficient represents the constraint term between legal and conflicting successive states. Indicates the first The flow of a single training sample is a real, reservable state quantity. Indicates the first The number of states that can be reserved for a single flow credential prediction of a training sample. This represents the minimum probability interval between a valid continuation state and a conflicting continuation state. Indicates the first The predicted probability of each training sample corresponding to the true legal state category. Indicates the first The predicted probability of the highest conflict state category corresponding to each training sample.

[0062] During this process, the training, validation, and test sets are separated according to production date, process route version, and compressor part identifier. The same event trajectory and its fault injection version are only included in one set. After each round of training, the positioning results for valid continuation, non-adjacent roles, inconsistent identifiers, unreservable credentials, and cross-workstation conflicts are checked. Training stops when the positioning results for any of these categories fail to improve after eight consecutive rounds of validation, and the parameter version with the smallest fluctuation in each category is retained. Model parameters are not modified during online operation. Newly added events that have completed on-site verification are entered into the offline candidate sample library in batches.

[0063] In some embodiments of this application, the role grid record vector is encoded in 32 dimensions, the identifier chain tail segment summary vector is encoded in 24 dimensions, the process route version encoding vector is encoded in 16 dimensions, and the first-time transfer credential status and role continuity compatibility are encoded in 8 dimensions after mapping. The splicing result enters a 64-dimensional role attention layer and a 64-dimensional gated state update layer, and the state output layer corresponds to 9 types of continuity states. The training batch size is set to 64, the initial learning rate is set to 0.001, the weight decay is set to 0.0001, and training is stopped when the validation metrics do not improve after 8 consecutive rounds. Each model version update freezes the training data range, process route version range, and role boundary version range, and online operation only calls the registered versions.

[0064] Furthermore, the model inference process first reads the model input vector of the current succession candidate, and then sequentially calculates the role record attention weight, candidate context vector, candidate update gate vector, candidate memory state, and succession state probability. The role succession conflict discrimination model cannot convert state edges outside the permitted process route into legal states, nor can it recover previously cancelled or reserved flow vouchers by other workstations. The set of reservable succession candidates retained after deterministic verification and model discrimination can be written as: , in, This represents the set of candidate successors that can be reserved after deterministic verification and model discrimination. This represents the complete set of continuation candidates formed after completing the identifier comparison and overlapping interval calculation. This indicates that the role succession compatibility quantity has entered the permission boundary of the set of reservable succession candidates. Indicates the first The predicted probability that each succession candidate belongs to the category of a valid succession state. Indicates the first The predicted probability that each successive candidate belongs to the highest conflict state category. This represents the lower bound of the probability that a valid continuation state will enter the set of reservable continuation candidates. This represents the upper limit of the probability that a conflict-prone succession state will enter the set of reservable succession candidates.

[0065] Based on this, the reserved workstations of a single-transfer credential are queried. Single-transfer credentials not registered in reserved workstations are bound to succession candidates, and single-transfer credentials registered in other workstations are written to the cross-workstation conflict record. During the query, the current reserved workstation, occupancy status, cancellation status, and last update time of the credential are read using the single-transfer credential identifier as the key. When no reserved workstation is registered and the credential is in an available state, an atomic write operation is used to simultaneously write the reserved workstation, succession candidate reference, credential succession sequence number, and reservation time. When other workstations have already been registered, the original reserved workstation is not overwritten; instead, the current workstation, the original reserved workstation, two candidate references, and the chain tail version are written to the cross-workstation conflict record.

[0066] It should be noted that after an exclusive reservation write fails, the status of the transfer credential is reread. If the failure is due to concurrent reservation, it is transferred to a cross-workstation conflict record. If the failure is due to storage transients, the current succession candidate is maintained and device licenses are restricted. If a transfer credential has been cancelled, the credential succession sequence number is less than the tail record, or the succession candidate set is empty, no new reservation record is generated. If the tail segment of the identifier chain, the process route version, or the role boundary version changes before the reservation is completed, the reservation is cancelled and the identifier chain verification is re-executed.

[0067] Thus, step S2 forms a succession verification chain consisting of role sequence arrangement, version validity verification, identification and overlapping interval comparison, role succession conflict discrimination model-assisted discrimination, and exclusive reservation of primary transfer credentials. It outputs succession candidates with expected reading role, credential succession sequence number, and unique reserved workstation, providing definite input for step S3 to establish control session and primary equipment license at the near end of the equipment.

[0068] In step S3, the core task is to match the succession candidates formed at the workstation entrance with the compressor part identifier, fixture identifier, and machining program identifier that actually exist near the equipment, and to establish a unique control session within the clamping boundary so that part replacement, fixture replacement, program switching, or repeated sessions that occur after the entrance verification cannot continue to obtain a single equipment permission.

[0069] In this embodiment, the device near-end reading channel begins data acquisition after the compressor part enters the fixture coverage area, and continuously records the compressor part identifier, fixture identifier, and machining program identifier until the clamping position signal is valid. The succession candidate provides the expected compressor part identifier, transfer fixture identifier, expected reading role, and credential succession number. The device near-end reading result provides the current compressor part identifier, current fixture identifier, current machining program identifier, continuous reading interval, and clamping position signal. The workstation controller establishes a control session only when the aforementioned object, credential succession number, and window boundary all correspond, and uses this control session to solidify the object combination used for subsequent device authorization.

[0070] Specifically, based on the continuous reading interval at the near end of the equipment, the compressor part identifier, fixture identifier, and machining program identifier are grouped into the same near-end window, and the near-end window is closed according to the clamping completion signal. Before grouping them into the same near-end window, the station controller corrects the acquisition time of the three identifiers according to a unified acquisition clock, and intersects the continuous reading interval with the effective time period of the clamping completion signal; after the clamping completion signal continuously reaches a fixed confirmation cycle, the near-end window is closed, and the reading event arriving after closure enters the next near-end window, avoiding irrelevant cross-reads after clamping completion from changing the current window.

[0071] Based on this, the continuously appearing read results in the near-end window are matched item by item with the succession candidates, and the succession sequence number of the credential is verified. When the compressor part identifier is replaced, the reservation of a transfer credential is cancelled. When the fixture identifier or machining program identifier does not match, a transfer credential is frozen. Item by item matching compares the compressor part identifier, fixture identifier, and machining program identifier respectively. The credential succession sequence number comparison is used to exclude historical succession candidates. When the compressor part identifier is replaced, the cancellation action first checks that the transfer credential is still in the reserved state and has not yet generated an executed permit, then clears the reserved workstation and increments the status version. When the fixture identifier or machining program identifier does not match, the freeze action retains the reserved workstation and credential succession sequence number, and writes the reason for resource incompatibility. The current near-end matching consistency of the equipment in the near-end window can be written as: , in, Indicates the first Device near-end matching consistency of a near-end window Indicates the first The first near-end window The consistency of each matching field with the subsequent candidate Indicates the first The contribution weight of each matching field to the near-end matching consistency of the device. Indicates the first The time drift of continuously reading results within a near-end window This represents the attenuation scale of the device's near-end matching consistency with time drift. Indicates the first The credential continuation sequence number read by the near-end window Indicates the first Each successor candidate carries a credential succession number.

[0072] Furthermore, when the near-end matching consistency reaches the permitted boundary and the three identifiers remain stable before the window closes, the primary transfer credential is changed from a reserved state to an occupied state after the read result matches the successor candidate. A control session number is generated, and the successor candidate, near-end window, and credential succession sequence number are associated with the control session number. Before the state transition, the reserved workstation, credential succession sequence number, and primary transfer credential status version are checked simultaneously. The transition stops if any field changes. The control session number is generated based on the workstation identity, the near-end window closing time, and the monotonically increasing sequence number within the workstation. It is unique within the same workstation. The control session record is written to the successor candidate reference, near-end window reference, credential succession sequence number, primary transfer credential identifier, fixture identifier, and machining program identifier.

[0073] In some embodiments of this application, a single device license allows only one control session to read the data, while other control sessions corresponding to the same single transfer credential remain in an unlicensed state. The validity of a single device license, while simultaneously verifying the device near-end matching consistency, the unique association of the single transfer credential across all matched control sessions, and the credential occupancy status, can be written as: , in, Indicates the first One device license validity period corresponding to each control session This indicates the permitted boundaries for near-end matching consistency of the device. Indicates the first The number of control sessions involved when a uniqueness verification is performed by a control session. In the uniqueness verification, the first A single transfer credential identifier associated with a control session. Indicates the first A single transfer credential identifier associated with a control session. This indicates the occupancy status of a single transfer credential corresponding to the current control session. It is set to 1 when the credential has been changed from the reserved state to the occupied state by the control session, and 0 otherwise.

[0074] During this process, a primary equipment permit is generated based on the control session number and written to the equipment controller. If the same primary transfer credential corresponds to another control session number or the read result changes, the unexecuted primary equipment permit is revoked, maintaining the occupancy status of the primary transfer credential. The primary equipment permit is written with the compressor part identifier, fixture identifier, machining program identifier, control session number, credential sequence number, permit generation time, and permit expiration boundary. Before actual startup, the equipment controller rereads the current fixture identifier and machining program identifier and verifies that the compressor part identifier remains present. When the same primary transfer credential has another control session number, any read result changes, or the permit exceeds the validity boundary, the equipment controller revokes the unexecuted primary equipment permit, and the primary transfer credential remains in an occupied state, awaiting review by the current control session or entering anomaly handling.

[0075] It should be noted that when the device's near-end reading channel temporarily loses the fixture identifier or processing program identifier, the credential occupation is not directly released. The near-end window retains the missing state and triggers the corresponding reading channel to replenish the credential. After the replenished credential obtains an identifier consistent with the successor candidate, the device's near-end matching consistency is recalculated. After the replenished credential obtains an inconsistent identifier, the transfer credential is frozen once. If the device controller has received a device license once but has not yet started, a program switch occurs, the license is revoked, and the processing program identifiers before and after the switch are recorded. Once the device has started, no new control session is generated, and subsequent events enter the execution window of step S4.

[0076] At this point, step S3 completes the secondary mapping between the succession candidate and the actual object at the near end of the equipment. The near-end window that meets the conditions of consistent compressor part identifier, fixture identifier, machining program identifier and credential succession sequence number is converted into a unique control session and primary equipment license. The cancellation or freeze status corresponding to object changes, resource mismatch and duplicate sessions is retained, providing a control basis for binding the actual machining cycle in step S4.

[0077] In step S4, the core task is to implement the primary equipment license into a verifiable processing cycle, and to limit the equipment start event, equipment completion event, and in-machine detection event to the same control session number, processing cycle identifier, and credential sequence number, thereby forming a process execution witness that can prove that the specified compressor part has completed the specified processing procedure.

[0078] In some embodiments of this application, after a device license is written to the device controller, the device controller reads back the current fixture identifier, machining program identifier, and compressor part presence status before startup, and compares them field by field with the control session record. After successful verification, the device controller assigns a machining cycle identifier to the actual machining operation, and makes the device start event, device completion event, and on-machine detection event carry the control session number, machining cycle identifier, credential sequence number, machining program identifier, and acquisition time, respectively. Thus, events returned by different interfaces can be attributed according to the same object relationship and the same machining cycle, thereby forming process execution witnesses.

[0079] Specifically, the equipment controller reads the one-time equipment license written in step S3, verifies the control session number, fixture identifier, processing program identifier, credential sequence number, and license validity boundary; after verification, it starts the current processing and assigns a processing cycle identifier. It can receive the processing cycle identifier returned by the equipment controller and associate it with the control session number and credential sequence number. The processing cycle identifier returned by the equipment controller, along with the license write sequence number, is written to the control session record; if a timeout occurs, the license is revoked, or the processing cycle identifier is already bound to another control session, the one-time transfer credential is retained and the current startup is terminated without opening an execution window.

[0080] Based on this, equipment start-up and equipment completion events are obtained from the equipment controller, and on-machine detection events are obtained from the on-machine detection interface. These three types of events are then written into the execution window corresponding to the processing cycle identifier. Equipment start-up and equipment completion events are obtained from the equipment controller's event interface, while on-machine detection events are obtained from the on-machine detection interface. Before entering the execution window, all three types of events are supplemented with the control session number, processing cycle identifier, credential sequence number, processing program identifier, and acquisition time. Events that cannot complete any of the bound fields are first written to the event set to be assigned and do not directly enter the current execution window.

[0081] Furthermore, an execution window is opened based on the device start event and stopped receiving new on-machine detection events based on the device completion event. The start and end times of the execution window are associated with the processing cycle identifier. After a device start event occurs, an execution window is created and its start time is recorded. After a device completion event occurs, the end time of the window is recorded. At this point, the execution window stops receiving new on-machine detection events, but retains already received events for subsequent attribution, sequence, and processing program identifier verification. When the same processing cycle identifier is repeatedly affected by a device start event, the earliest valid device start event is retained, and subsequent events are written to the conflict event set. If a device completion event occurs earlier than a device start event, or if a device completion event occurs again after the window has stopped collecting data, the existing execution window boundaries are not changed.

[0082] In some embodiments of this application, on-device detection events within the execution window are arranged according to their acquisition time. The control session number and credential sequence number recorded in the on-device detection event records are compared, and on-device detection events that do not match are written into a conflict event set. The original acquisition time and interface reception time of the on-device detection events are retained during the arrangement; out-of-order interface reception does not change the acquisition time order. Events whose control session number or credential sequence number does not match the current execution window are written into the conflict event set, and their original processing cycle identifier is retained for subsequent determination of whether they belong to adjacent processing cycles.

[0083] Specifically, based on the acquisition time of the equipment completion event, adjacent detection confirmation events are determined from the on-machine detection events that have not been written into the conflict event set, and the processing procedure identifier of the detection confirmation event is compared with the processing procedure identifier of the primary equipment license record. When selecting a detection confirmation event from the on-machine detection events that have not been written into the conflict event set, the processing procedure identifier and the credential sequence number must be consistent, and the absolute interval between the acquisition time of the detection event and the acquisition time of the equipment completion event must be minimized. The index of the detection confirmation event adjacent to the equipment completion event and with the same object in the current execution window can be written as: , in, Indicates the first The index of detection confirmation events adjacent to and identical to the device completion event in the execution window. Indicates the first The set of detected events that were not written into the conflict event set within each execution window. Indicates the first The time of data collection for each detection event. Indicates the first The device completes event acquisition time corresponding to each execution window. Indicates the first The processing procedure identifier code of each detection event record. Indicates the first The processing procedure identifier code of the primary equipment license record associated with each execution window. Indicates the first The credential sequence number of each detection event record. Indicates the first The credential sequence number associated with each execution window.

[0084] During this process, when the processing procedure identifiers of the detection and confirmation events are consistent, the device start event, device completion event, and detection and confirmation event are confirmed as the valid event set for the current execution window. When the detection and confirmation event is missing or the processing procedure identifiers are inconsistent, the execution window is marked as an event gap. Subsequently, the control session number, processing cycle identifier, and credential sequence number of the three types of event records are checked. Events with consistent content are formed into an event sequence according to the collection order, while events with inconsistent content are written into a conflict event set. After checking each field of the three content, a consistency value for binding the three types of events is formed. The control session number, processing cycle identifier, and credential sequence number each correspond to an independent binding field. If any field is inconsistent, even if the collection times are adjacent, the event cannot enter the event sequence. The consistency value for binding the three types of events within the current execution window can be written as: , in, Indicates the first Consistent binding of the three types of events within an execution window Indicates the first The device startup event collection time corresponding to each execution window Indicates the first The selected detection and confirmation event acquisition time for each execution window. Indicates the first The device completes event acquisition time corresponding to each execution window. Indicates the first The first execution window Consistency of the bound field, the first Each binding field corresponds to the control session number, processing cycle identifier, and credential continuation sequence number, respectively.

[0085] Based on this, the event sequence is arranged according to the acquisition time of the equipment start event, the detection confirmation event, and the equipment completion event. No equipment start event or equipment completion event belonging to another control session may be inserted between the equipment start event and the equipment completion event. A process execution witness is formed when the three types of events are bound consistently, the event sequence is established, the processing program identifier is consistent, and the detection confirmation event is within a fixed detection boundary. The valid quantity of the process execution witness formed in the current execution window can be written as: , in, Indicates the first The effective quantity of process execution witnesses formed by each execution window This indicates the attenuation scale of the time interval between the process execution witnessing event and the equipment completion event. Indicates the first The processing procedure identifier code of the selected detection confirmation event record in the execution window.

[0086] In some embodiments of this application, process execution witnesses are formed based on the successive verification results of the event sequence. When the effective quantity of the process execution witness reaches the confirmation boundary, the execution window completes the verification and archives it, and simultaneously writes the process execution witness; when events are missing, successive relationships are not established, processing program identifiers are inconsistent, or detection confirmation events exceed fixed detection boundaries, the execution window archives the event in an event gap state and retains the reference to the conflicting event, the primary transfer credential continues to remain in an occupied state, and the equipment exit remains locked.

[0087] It should be noted that when the device event interface is temporarily interrupted, the local event cache only records the original events and the interface sequence, and does not automatically form process execution witnesses. After the interface is restored, the events are reassigned according to the control session number, processing cycle identifier, and credential sequence number. After obtaining the missing events through supplementary certification and verifying the sequence, the effective quantity of process execution witnesses is recalculated. If supplementary certification still cannot close the current event sequence, the execution window maintains the event gap state, and the reuse of events from other execution windows is prohibited.

[0088] At this point, step S4 completes the object binding and timing closure between a single equipment license, processing cycle, and three types of field events, converting the event sequence that can prove that the specified control session actually completed the processing into process execution witnesses, while retaining the frozen boundaries corresponding to event gaps and conflict events, providing factual basis for step S5 to execute the identifier segment writing, voucher cancellation, and chain tail update by submission batch.

[0089] In step S5, the core task is to convert the process execution witness into the process confirmation identifier segment writing, one-time flow voucher cancellation, and identifier chain tail segment update under the same submission batch, and to determine whether the submission process forms a complete closed loop through the payload, operation sequence, voucher continuation number, and chain tail version of the three types of receipts.

[0090] In some embodiments of this application, the submission controller first locates the compressor part identifier, control session number, processing cycle identifier, and document sequence number based on the process execution witness, then reads the current tail version of the identifier chain segment and freezes it as the submission base version; subsequently, it associates the process confirmation identifier segment writing, one-time flow document cancellation, and identifier chain tail segment update triggered by the same process execution witness according to the submission batch number. Any operation retry continues to use the original submission batch number, original operation sequence number, and original submission load to prevent the retry process from forming another set of submission relationships; finally, it obtains the write receipt, cancellation receipt, and update receipt.

[0091] Specifically, a submission batch number and submission sequence are generated based on the process execution witness. The write sequence number, verification sequence number, and update sequence number are obtained from the submission sequence, and the tail version of the chain segment is recorded as the submission base version. The submission batch number is generated based on the workstation identity, control session number, processing cycle identifier, and monotonic sequence number within the batch. The submission sequence always includes write operations, verification operations, and update operations. The write sequence number, verification sequence number, and update sequence number are assigned one-time according to the submission sequence. Retry operations retain the original sequence number to avoid conflicting operation sequences within the same submission batch due to network retries.

[0092] Based on this, the control session number and processing cycle identifier are categorized into session items, and the credential continuation sequence number and continuation candidate are categorized into continuation items. Session items and continuation items are combined to form the commit payload. The session item records the control session number, processing cycle identifier, and process execution witness reference, while the continuation item records the credential continuation sequence number, continuation candidate reference, expected read role, and first-time flow credential identifier. The commit payload also retains the process route version, chain tail version, and processing program identifier, enabling the three types of operations to compare the content of the same object.

[0093] Furthermore, the process confirmation identifier segment is written according to the submitted batch number, and a write receipt is obtained recording the submitted batch number, write sequence number, and submitted load. The process confirmation identifier segment writes the compressor part identifier, transfer tool identifier, fixture identifier, machining program identifier, control session number, machining cycle identifier, document continuation sequence number, process execution witness, and reference to the previous identifier chain tail segment; when the same submitted batch number and write sequence number are received repeatedly, the original write receipt is returned, and a new process confirmation identifier segment is not written again.

[0094] In some embodiments of this application, a single circulation credential is revoked based on the submission batch number. After confirming that the credential continuation sequence number matches, a revocation receipt recording the submission batch number, revocation sequence number, and submission payload is obtained. The revocation operation first verifies that the circulation credential is in an occupied state, the occupied workstation matches the current workstation, and the credential continuation sequence number matches the submission payload. After successful verification, the credential status is changed to a revoked state. If the same submission batch number and revocation sequence number are received repeatedly, the original revocation receipt is returned. If a circulation credential has already been revoked by another submission batch, a continuation conflict is written.

[0095] During this process, once the tail version matches the submission baseline version, the tail segment is updated to the process confirmation segment, and an update receipt is obtained recording the submission batch number, update sequence number, submission payload, the tail version before the update, and the tail version after the update. The update operation uses the submission baseline version for comparison writing. After the tail versions match, the tail segment is pointed to the current process confirmation segment, forming a new tail version. If the tail versions do not match, the current tail segment is not overwritten. The update receipt simultaneously records the tail version before the update, the tail version after the update, the submission batch number, the update sequence number, and the submission payload summary.

[0096] Further, the write receipt, cancellation receipt, and update receipt are verified, and a submission status is generated based on the verification results. Before verification begins, the three types of receipts are grouped into a receipt set. Receipts not related to the current submission are filtered out by submission batch number, and duplicate receipts are filtered out by receipt type and corresponding sequence number. Receipts not related to the current submission are filtered out by submission batch number, and duplicate receipts are filtered out by receipt type and corresponding operation sequence number. The filtered records retain the original receiving time and source interface, without changing the current receipt set.

[0097] Specifically, the comparison of session items and succession items in the receipt set is used to identify any inconsistency as a load conflict. The order of write, cancellation, and update numbers is compared, and duplicate or reversed numbers are identified as operation number conflicts. The verification receipt is compared with the credential succession number of the process execution witness; the smaller former is identified as a historical replay, and the larger former is identified as a succession conflict. The comparison of session items and succession items combines submitting a load summary with field-by-field review. Inconsistent summaries directly result in load conflicts; identical summaries but differences in individual fields still result in load conflicts. Duplicate or reversed operation numbers result in operation number conflicts. A credential succession number less than the corresponding number in the process execution witness results in a historical replay, while a greater number results in a succession conflict.

[0098] Based on this, the pre-update tail version of the update receipt is compared with the submission baseline version. Inconsistencies in the pre-update tail version, operation sequence number conflicts, or historical replays are recorded as timing conflicts, while successive conflicts are recorded as payload conflicts. Inconsistencies in the pre-update tail version, operation sequence number conflicts, or historical replays all indicate that the submission operation did not close along the same time sequence, and are uniformly written as timing conflicts; successive conflicts indicate that the corresponding object or process of the voucher has shifted, and are written as payload conflicts. The operation sequence, pre-update tail version, and submission payload summary of the three types of receipts together form the consistency quantity for the three types of receipts, which can be written as: , in, Indicates the first The consistency of the three types of receipts in each submission batch Indicates the first The write sequence number of the write process confirmation identifier segment for each batch of submissions. Indicates the first The verification sequence number of each batch of verification vouchers submitted for verification. Indicates the first The update sequence number of the tail segment of the batch update identifier. Indicates the first The previous tail version of the commit batch update receipt record. Indicates the first Each commit batch reads the commit baseline version. Indicates the first In the first submission batch Submission payload summary of the acknowledgment record, number The corresponding receipts are written to the receipt, the cancellation receipt, and the update receipt in sequence. Indicates the first The submission payload summary formed by each submission batch.

[0099] In some embodiments of this application, when the receipt set has no receipt gaps, load conflicts, or timing conflicts, the receipt set is recorded as a complete submission state; when any issue exists, the receipt set is recorded as an incomplete submission state, and the corresponding issue is written into the submission state. The complete and incomplete submission states are determined jointly according to the three types of receipt consistency and the receipt type set, and can be written as follows: , in, Indicates the first The commit status formed by each commit batch Indicates the first Each batch of submissions filters out the set of receipt types to retain, excluding the current receipt and duplicate receipts.

[0100] Furthermore, a complete submission status includes the submission batch number, process confirmation identifier reference, one-time cleared transfer voucher reference, updated chain tail version, and three types of receipt references; an incomplete submission status includes receipt gap, load conflict, timing conflict, continuation conflict, and corresponding original receipt reference. If only a receipt gap exists, completed operations are not rolled back; subsequent steps (S6) verify the equipment execution facts and decide whether to perform the missing operation. If a load conflict or timing conflict exists, subsequent supplementary execution is stopped to prevent erroneous loads from continuing to advance to the chain tail.

[0101] It should be noted that after a storage interface transient failure, the same submission batch can retry operations that did not receive a receipt, but the operation sequence number must not be reassigned or the submission payload changed. When the process confirmation identifier segment has been written but the reversal receipt is missing, the recovery process first reads the current status of the transfer document; if the transfer document has already been reversed by this submission batch, the original receipt is retrieved; if the transfer document is still in an occupied state, the reversal is performed; if the document has been reversed by another batch, a payload conflict will occur.

[0102] Thus, step S5 establishes a submission control relationship with process execution witness as the submission starting point, submission baseline version as the concurrency boundary, and three types of receipts as closed-loop evidence. This relationship can distinguish between complete submissions, gaps where receipts can be retrieved, load conflicts, timing conflicts, and historical replays, providing an executable submission state for step S6 to control the equipment exit lock state.

[0103] In step S6, the core task is to convert the submission status formed in step S5 into device exit lock, recovery supplementary execution, isolation hold or unlock confirmation, and ensure that the device exit status readback is consistent with the current submission batch, control session and identifier chain tail segment, so as to prevent compressor parts from leaving the device prematurely when the digital submission is not closed.

[0104] In some embodiments of this application, the equipment exit remains locked by default after the current compressor part enters the controlled workstation. The exit controller does not use a separate write receipt, a separate verification receipt, a separate update receipt, or only the sending of an unlock command as the basis for release; only when the submission batch number, session item, succession item, operation sequence number, and submission payload of the three types of receipts are complete and consistent, the update receipt record's previous chain tail version is consistent with the submission baseline version, and the equipment exit status readback corresponds to the unlock command, is an unlock confirmation written and the equipment exit lock status released; when any condition is not met, the equipment exit lock status is maintained and the transfer voucher is frozen once.

[0105] Specifically, the device's exit lock status is associated with the submission batch number to generate an exit lock sequence number. The complete or incomplete submission status is written to the exit control record. The exit lock sequence number monotonically increases within the same device exit. The exit control record also includes the submission batch number, control session number, processing cycle identifier, credential continuation sequence number, submission status, unlock command status, and exit status readback. When a new submission batch enters the same device exit, it does not overwrite the exit control record that is still in a locked state.

[0106] Based on this, the device exit is kept locked and a single transfer document is frozen according to the incomplete submission status. Subsequent device licenses corresponding to the same compressor part identifier are rejected, and the exit lock number is written to the rejection record. The rejection record retains the compressor part identifier, the single transfer document identifier, the control session number, the exit lock number, and the rejection reason. When the device controller receives a subsequent device license corresponding to the same compressor part identifier, it first searches for unclosed rejection records. If an unclosed rejection record exists, the license is rejected.

[0107] Furthermore, based on the submission batch number, a receipt set and a process execution witness are obtained. The control session number, processing cycle identifier, and credential sequence number in the receipt set are compared with the corresponding content in the process execution witness. During the comparison, the receipt set is located by the submission batch number, and the original execution window is located by the process execution witness reference. The control session number, processing cycle identifier, and credential sequence number are checked respectively. If a certain type of receipt is missing from the receipt set but the existing receipts match the process execution witness, recovery is possible. If any object in the existing receipts is different, supplementary execution is not allowed.

[0108] In some embodiments of this application, a processing cycle identifier and a credential sequence number are obtained from the equipment controller. These two pieces of information are compared with the corresponding information in the process execution witness. Any discrepancy is recorded as an execution conflict. The processing cycle identifier and credential sequence number returned by the equipment controller are used to prove that the actual equipment still corresponds to this submitted batch. Any discrepancy is written as an execution conflict, and the current value of the equipment controller, the process execution witness value, and the comparison time are recorded. Execution conflicts cannot be eliminated by retrieving a receipt, and the relevant compressor parts remain at the equipment outlet awaiting isolation and disposal.

[0109] Specifically, when all aspects are consistent and only a receipt gap exists, the corresponding problem is recorded as a recoverable gap. The device exit is kept locked and the first-time transfer document is frozen. The missing operation is then performed using the submission batch number, and the new receipt is written to the receipt set. Before the supplementary execution, the tail segment of the identifier chain, the status of the first-time transfer document, and the original receipt of the completed operation are read again to confirm that the completed operation has not been changed by other submission batches. The new receipt obtained after the missing operation is completed uses the original submission batch number and the original operation sequence number, and the three types of receipt verification in step S5 are re-executed. Only after a complete submission status is formed will the unlock branch be entered.

[0110] During this process, when load conflicts, timing conflicts, or execution conflicts exist, the corresponding issues are recorded as isolation conflicts. The missing operations are rejected for re-execution, the equipment exit remains locked, and the primary transfer document is frozen. The compressor part identifier and submission batch number are written into the isolation record. The isolation record includes the compressor part identifier, submission batch number, control session number, processing cycle identifier, document sequence number, conflict type, and original receipt reference. The equipment exit remains locked, the primary transfer document remains frozen, and any re-execution requests and new equipment permits are rejected until a clear resolution is reached through on-site verification.

[0111] Furthermore, the recovery or isolation of incomplete submission status is jointly determined by receipt gaps, load conflicts, timing conflicts, and execution conflicts. The current incomplete submission status handling status corresponding to the equipment exit control record can be written as: , in, Indicates the first The status of handling incomplete submissions corresponding to the exit control records of each device. Indicates the first The status quantity indicating whether each submission batch only has a receipt gap is the receipt gap status quantity. Indicates the first A load conflict status quantity indicating whether there are load conflicts in each submission batch. Indicates the first The timing conflict status quantity indicating whether there are timing conflicts in each submission batch. Indicates the first The execution conflict status quantity indicates whether there is an execution conflict in each submitted batch.

[0112] In some embodiments of this application, the process confirmation identifier segment of the update receipt record is compared with the process confirmation identifier segment of the identifier chain tail record based on the complete submission status. If they match, an unlocking command containing the exit lock sequence number and control session number is generated. The comparison uses the process confirmation identifier segment summary of the update receipt record and the process confirmation identifier segment summary of the currently recorded identifier chain tail. If the two summaries match, the updated chain tail version of the update receipt is then compared with the current chain tail version. After successful verification, an unlocking command is generated, which simultaneously carries the exit lock sequence number, control session number, submission batch number, and target exit status.

[0113] Based on this, the unlock command is written to the equipment controller. The controller receives the exit status readback containing the exit lock sequence number and control session number. It compares the two items in the exit status readback with those in the unlock command. If both items match, an unlock confirmation is written and the equipment exit lock is released. If either item does not match, a lock confirmation is written and the equipment exit lock remains active. After executing the unlock command, the equipment controller returns the exit lock sequence number, control session number, submission batch number, and actual exit status. The exit control record compares the readback content with the unlock command field by field, verifying that the submission status is still a complete submission status and that the process confirmation identifier segment summary has not changed. The equipment exit unlock confirmation quantity corresponding to the current equipment exit control record can be written as: , in, Indicates the first Each device's exit control record corresponds to the device's exit unlock confirmation quantity. Indicates the first Summary of the process confirmation identifier segment of the update receipt record for each submitted batch. This indicates the summary of the process confirmation segment currently recorded at the end of the identification chain. Indicates the first The exit status readback vector obtained from the exit control record of each device. Indicates the first Each device's exit control record is written into the device controller's unlock command vector.

[0114] It should be noted that if the device export status readback times out, the export lock sequence number is inconsistent, the control session number is inconsistent, or the actual export status is still locked, a lock confirmation is written and the device export lock status is maintained, and a new export lock sequence number is not generated repeatedly; when all readback contents are consistent and the device export unlock confirmation quantity is valid, an unlock confirmation is written to release the device export lock status, and the unlock confirmation is associated with the submitted batch number, the tail segment of the identification chain, and the already verified once-transfer voucher.

[0115] At this point, step S6 completes the closed-loop control between the submission status, equipment execution facts, the tail segment of the identification chain, and the equipment exit status readback. This allows submissions with only a receipt gap and consistent object relationships to be restored using the original submission batch. It also isolates load conflicts, timing conflicts, or execution conflicts. Furthermore, the equipment exit lock status is released only when all three types of receipts and the exit readback are consistent, thereby ending the controlled station flow of the compressor parts.

[0116] In summary, this method incorporates the on-site identification relationships of compressor parts, transfer tools, fixtures, and machining programs into a continuous identification chain verification. It also uses process route version, one-time transfer voucher, process execution witness, submission receipt, and exit status rereading to jointly constrain the transfer status, achieving continuous consistency between on-site bearing relationships, equipment processing status, and digital records. This method can suppress erroneous authorization and release caused by cross-reading, object replacement, out-of-order continuation, replay of historical vouchers, and incomplete submissions, thereby improving the stability of cross-station transfer status correspondence, the certainty of anomaly blocking, the credibility of process confirmation, and the reliability of the exit release closed loop.

[0117] In some embodiments of this application, to illustrate the coordination between various state objects, computational relationships, and abnormal branches, a compressor parts transfer error prevention method based on identifier chain verification is configured on an automated production line for variable displacement automotive air conditioning compressor pistons. This production line is equipped with a loading station, a turning station, a dimensional inspection station, and corresponding equipment exits. The transfer fixture uses a single-bearing-position positioning plate. The piston, transfer fixture, turning fixture, inspection fixture, and machining program are each assigned machine-readable identifiers. For example, the piston identifier is P-VA-260730-041. It performs a complete transfer task from the loading station to the turning station, completing turning, transferring to the dimensional inspection station, and leaving the inspection station. The transfer fixture identifier is T-0428, the process route version is R-17, and the role boundary version is RB-08.

[0118] A complete implementation process may include the following steps: Step 1: Organize historical event trajectories and complete offline training of the role succession conflict discrimination model. Training samples are selected from production records that have undergone on-site verification and can simultaneously trace role grid records, identifier chain tail segments, process route versions, and first-time transfer document status, forming a total of 18,000 event trajectories. Training, validation, and test sets are separated according to production date, process route version, and compressor part identifier, with 12,600, 2,700, and 2,700 trajectories configured respectively. The same original trajectory and its fault-injected version are only included in one set. Nine types of state samples include 8,400 valid successions, 1,200 non-adjacent roles, 1,000 invalid role boundary versions, 1,000 invalid process route versions, 1,400 inconsistent compressor part identifiers, 1,200 inconsistent transfer document identifiers, 1,100 insufficient overlapping intervals, 1,400 unreservable documents, and 1,300 cross-workstation conflicts, ensuring stable coverage of low-frequency conflict states in each training batch.

[0119] Step 2: Construct model inputs and relational hierarchies adapted to this production line. For each succession candidate, read a 32-dimensional role raster record vector, a 24-dimensional identifier chain tail summary vector, and a 16-dimensional process route version encoding vector. Map the reservable state quantities of a single transfer credential to the role succession compatibility quantities into an 8-dimensional state vector, forming an 80-dimensional model input in a fixed field order. The role raster record vector contains shared confidence quantities, vehicle occupancy validity quantities, reading role, role boundary version, continuous reading window start and end boundaries, short-term read missing states, serial read states, and original read event references. The identifier chain tail summary vector contains the most recently confirmed compressor part identifier, transfer tool identifier, end role, process route version, credential succession sequence number, and chain tail version. The process route version encoding vector contains the permission relationship from the end role to the expected reading role, as well as fixture, machining program, and minimum migration time constraints.

[0120] The role attention layer projects role raster records associated with the same continuation candidate into a 64-dimensional feature space. It allocates contributions based on persistence, missing states, conflict states, and their coupling relationship with candidate inputs, preventing short-term cross-reads from overlapping stable shared records in adjacent read channels. The aggregated candidate context vectors and the 80-dimensional model input enter a 64-dimensional gated state update layer. This layer retains the preceding memory formed by the tail segment of the identifier chain while restricting the writing of role boundary version failures, process route version failures, or unreservable credential states into valid continuation directions. The output layer sets nine state positions and simultaneously outputs the number of reservable states for a single-transfer credential. The process route permission mask is located before the output layer; state edges not permitted by valid process routes cannot enter valid continuation positions, and single-transfer credentials that have been cancelled or reserved by other workstations cannot be recovered by the model.

[0121] Step 3: Train the model in three stages. The first stage fixes the process route permission mask and the exclusive rule for one-time transfer credentials, updating only the role attention layer, gating state update layer, and state output layer. The second stage adds continuous read boundary jitter, missing single role records, adjacent channel cross-reading, and overlapping interval shortening samples. The third stage adds process route version switching, low-frequency piston models, historical credential replay, and cross-workstation concurrent reserved samples. Data augmentation uses the on-site collection cycle as the boundary, changing only one object relationship each time while keeping the other roles, identifiers, versions, and credential states unchanged, thus ensuring that conflict labels correspond to clear physical causes. The training batch size is 64, the initial learning rate is 0.001, and the weight decay is 0.0001. After each round, the localization results for the nine states are statistically analyzed, and training stops when the results do not improve after eight consecutive rounds of verification.

[0122] In addition to the constraints of classification and process route structure, the training objective further constrains the probability interval between the state of a single transfer document, the legal state, and the highest conflict state, which can be written as: , In this embodiment, the training coordination coefficient Set to 0.25, training coordination coefficient Take 0.15, minimum probability interval Set the value to 0.30. The classification and process route structure constraints obtained from the 67th round of verification. The value is 0.061. The status constraint term for a single transaction document is converted to 0.012, and the probability interval constraint term is converted to 0.013. Substituting these values ​​into the model training target... The value was 0.086; no continuous improvement was observed from rounds 68 to 75, therefore the parameters for round 67 were frozen and registered as model version RCM-202607-04. Online operation only reads the parameters of this model version and does not update the weights during production; new trajectories that have completed on-site verification are subsequently added to the offline candidate sample library in batches according to date.

[0123] Step four: Establish the physical co-occupancy relationship between the piston and the transfer device at the loading position. After the positioning plate T-0428 enters the loading position, the positioning pin arrival signal and the load-bearing position photoelectric signal are simultaneously valid, and the loading position reading channel opens the continuous reading window. The piston P-VA-260730-041 stably reaches 2.70s, and the positioning plate T-0428 stably reaches 2.78s. The overlap length of the two continuous reading intervals is 2.70s, and the maximum time deviation at the start and end boundaries is [not specified]. The effective occupancy time of the vehicle is 0.04s. Set to 1, attenuation scale Let's take 0.40s. The co-occurrence confidence value for the current continuous read window can be written as: , After substituting the above input values, the total confidence level is... The value is 0.8916, which is higher than the fixed boundary of 0.82 for the loading position. Therefore, the loading position retains two identifiers, the continuous reading interval, the vehicle occupancy status, and the original reading event, and writes the co-occupancy record to the role grid position corresponding to the loading reading role. Before the window closed, the identifier of the adjacent positioning disk T-0429 had one acquisition cycle. This reading did not form an interval that continued together with the vehicle occupancy status; it was written to the serial read status and the original channel reference was retained, without changing the main co-occupancy record of T-0428 and P-VA-260730-041.

[0124] It should be noted that the "Document Continuation Sequence Number" and "Chain Tail Version" in this complete implementation process are data fields written during the identification chain control process and updated with the flow status. The following 16, 17, and 18 are example values ​​for the Document Continuation Sequence Number, where 16 represents the document continuation order recorded in the identification chain tail segment before the turning station continuation; 17 represents the current continuation order of the turning station generated according to the process confirmation order based on the aforementioned document continuation order; and 18 represents the next continuation order generated by the dimensional inspection station after the turning process confirmation. 304, 305, and 306 are example values ​​for the Chain Tail Version, where 304 represents the current version status of the identification chain tail segment before the turning process submission; 305 represents the chain tail version after the turning process confirmation submission; and 306 represents the chain tail version after the dimensional inspection process confirmation submission. These values ​​are used to illustrate the progressive relationship between the document continuation order and the identification chain tail segment version as the process flows.

[0125] Step 5: Verify the identification chain at the turning station entrance and exclusively reserve one transfer document. After obtaining the valid co-occupancy records P-VA-260730-041 and T-0428 at the turning station entrance, form a role sequence according to the role boundary version, continuous reading window start time, and record writing order; read the end role in the tail segment of the identification chain, the current example value of the process route version R-17, and the document continuation sequence number is 16, and the current example value of the tail version is 304. Process route version R-17 limits the next role to the turning station entrance reading role, the current role boundary version is in the activation cycle, and the two identifiers are consistent with the most recently confirmed value at the tail. The adjacent permission quantity from the end role to the first valid role is 1, the valid quantity of both types of versions is 1, and the role sequence deviation is... Take 0.04 as the attenuation scale. Taking 0.50, the character continuity compatibility can be written as: , Character continuity compatibility after substitution The value is 0.9231, higher than the permitted boundary of 0.85. The current role grid record, identifier chain tail segment summary, process route version code, first-time transfer credential reservation status, and this calculation result are input into model version RCM-202607-04. The output is a valid continuation probability of 0.963 and a maximum conflict state probability of 0.008; the valid continuation probability reaches 0.90 and the maximum conflict state probability is lower than 0.08. Subsequently, an atomic write is performed using the first-time transfer credential identifier as the key, simultaneously registering the turning station, current continuation candidate, credential continuation sequence number, and reservation time. The example value for the current credential continuation sequence number is 17. The readback status version only increments by 1, the reserved station remains the turning station, and exclusive reservation is completed.

[0126] Step Six: Verify the workpiece at the near end of the turning equipment and perform machining. After the piston enters the area covered by the turning fixture, continuously read the compressor part identifier, fixture identifier, and machining program identifier, and close the near-end window with the valid edge of the clamping position signal. The first read fixture identifier is F-TN-17, which conforms to process route version R-17, but the machining program identifier is PRG-VA-16, which is inconsistent with the PRG-VA-17 required by the succession candidate; the primary transfer document retains the reserved station and document succession number, the status is changed to frozen, and the reason for the machining program incompatibility is recorded. After loading PRG-VA-17 at the operating end, the near-end window is re-established. The three identifiers remain stable for 1.20s before clamping position. The example value of the document succession number this time is 17, and this document succession number is consistent with the succession candidate and the current value of the primary transfer document.

[0127] After resolving the resource mismatch, the primary transfer credential is changed from reserved to occupied, generating a control session number CS-TN-20260730-0081. The succession candidate, near-end window, credential succession number, fixture identifier, and machining program identifier are written to the control session record. A primary equipment permit is generated based on the control session number. The turning equipment accepts the permit under the following conditions: the inlet lock is closed, the clamping signal is valid, the machining cycle has not yet started, and the permit has not been revoked. The machining cycle identifier CY-TN-0081 is returned. The equipment start event opens the execution window; the on-machine detection interface returns a piston outer diameter detection event; and the equipment completion event closes the execution window. All three types of events carry the control session number CS-TN-20260730-0081, the machining cycle identifier CY-TN-0081, and the credential succession number (example value is 17). The machining program identifier is PRG-VA-17.

[0128] The detection confirmation event occurs 0.36 seconds before the device completion event. For all three types of events, the control session number, processing cycle identifier, and credential sequence number are consistent across all fields, and the event binding consistency is maintained. Set to 1, time decay scale Take 1.50 seconds and verify that the processing procedure identifier of the event matches the primary equipment permission record. The valid quantity of the process execution witness formed in the current execution window can be written as: , Substitute into the post-process execution witness effective quantity The value is 0.7866, which is higher than the confirmation boundary of 0.75. Based on this, the execution window saves the original references, test results, and start and end times of the equipment start event, test confirmation event, and equipment completion event, forming a witness to the turning process execution; the one-time transfer voucher remains in an occupied state, waiting for the process confirmation and submission to complete.

[0129] Step 7: Complete the three submissions for the turning operation and transfer to the dimensional inspection station. Read the current example value of the tail version as 304 and use this tail version as the submission base version. Generate a submission batch number BT-TN-0081 based on the compressor part identifier, control session number, machining cycle identifier, and submission start time. Assign the write sequence number 31021, cancellation sequence number 31022, and update sequence number 31023 in a fixed order. After writing the operation confirmation identifier segment, obtain the write receipt. Verify the occupancy status of the transfer document, control session number, and document continuation sequence number to obtain the cancellation receipt. Then, perform a comparison write using the submission base version, pointing the tail segment identifier to the current operation confirmation identifier segment. The update receipt arrives with a 0.80s delay due to network jitter. During the recovery process, the original submission batch number and update sequence number are used to query the target status. It is found that the example value of the tail version has increased from 304 to 305 and the submission payload summary is consistent. Therefore, only the existing update receipt is retrieved, and the tail segment identifier is not updated again.

[0130] The submission batch number, session item, succession item, and submission payload summary of the three types of receipts are consistent. The operation sequence number is kept between 31021 and 31022, and between 31022 and 31023. The voucher succession sequence number in the verification receipt is equal to the value of the turning process execution witness record. The version at the end of the chain before the update is equal to the submission baseline version. The receipt set is recorded as a complete submission status. The equipment exit control record generates a unique exit lock sequence number and writes the unlock command to the turning equipment controller. The exit status readback carries the same exit lock sequence number, control session number, and submission batch number, and shows that the actual exit status has been unlocked. The piston after turning enters the dimensional inspection station entrance with the positioning plate T-0428.

[0131] Step 8: At the dimensional inspection station, complete the second continuation along the updated identifier chain. The dimensional inspection station entry reads P-VA-260730-041 and T-0428, establishing a new continuous reading window and role grid record. The tail of the identifier chain now points to the turning process confirmation identifier segment. The example value of the credential continuation sequence number becomes 17. Process route version R-17 limits the next role to the dimensional inspection station entry reading role. The role continuation conflict discrimination model outputs a valid continuation probability of 0.978 and a maximum conflict state probability of 0.005, exclusively reserving a new credential continuation sequence number with an example value of 18. Subsequently, the inspection fixture identifier F-IN-08 and the inspection program identifier PRG-IN-17 are read, establishing the control session number CS-IN-20260730-0026 and obtaining the machining cycle identifier CY-IN-0026. The equipment start event, dimensional inspection confirmation event, and equipment completion event form the same execution window, and the inspection result is written to the dimensional inspection process execution witness.

[0132] The batch number BT-IN-0026 was submitted to the dimensional inspection process confirmation segment, the primary transfer document was cancelled, and the tail segment of the identification chain was updated. All three types of receipts were complete and consistent in content, and the example value of the tail version increased from 305 to 306. After receiving the unlock command, the dimensional inspection equipment controller returned a consistent exit lock sequence number and control session number, and the equipment exit was unlocked. Subsequently, the network interface received another historical reservation request for the turning station corresponding to the example value of the document continuation sequence number 17. The current tail segment of the identification chain has recorded a new document continuation sequence number with an example value of 18, and the original primary transfer document has been cancelled. This request was written into the history replay state, and no new continuation candidate, control session number, or equipment license was formed.

[0133] Through the complete implementation process described above, this method incorporates the continuous co-occupancy relationship between compressor parts and transfer tools, cross-station role succession, equipment near-end resource verification, actual processing events, and three types of submission receipts into the same identification chain control process. This ensures that the on-site carrying objects, processing resources, process execution results, and digital status continuously correspond, and solidifies serial reading, object replacement, program mismatch, historical document replay, and local receipt gaps into locationable control states. This reduces the risk of incorrect parts combinations obtaining equipment permission or leaving the station before completing the closed loop in continuous flow of turning and dimensional inspection, reduces reliance on manual barcode scanning and experience verification, and improves cross-station error prevention stability, process confirmation reliability, anomaly recovery certainty, and exit release reliability, thereby improving the quality controllability and production safety of continuous processing of compressor parts.

[0134] like Figure 2As shown, another embodiment of this application provides a compressor parts transfer error prevention system based on identifier chain verification. This system includes functional modules corresponding to each step of the above-described compressor parts transfer error prevention method based on identifier chain verification. The specific data processing procedures for each module can be found in the above-described method embodiments.

[0135] Based on the same concept, another embodiment of this application provides a compressor parts transfer error prevention system based on identifier chain verification, comprising: The co-occupancy writing module 1 is used to read compressor part identifiers and transfer fixture identifiers at loading positions and workstation entrances, form co-occupancy records according to continuous reading intervals and fixture occupancy status, and write them to the role grid according to the read role and role boundary version. The identifier verification module 2 is used to verify the identifier chain based on the role grid, the tail of the identifier chain, and the process route version, forming a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserving one flow credential. Session authorization module 3 is used to read compressor part identifier, fixture identifier and machining program identifier at the near end of the equipment, match the reading results with the success candidates, generate a control session number and associate it with a device authorization and a transfer credential; Witness Formation Module 4 is used to start processing according to a single equipment license, and bind the equipment start event, equipment completion event and in-machine detection event to the control session number, processing cycle identifier and credential sequence number to form a process execution witness; The confirmation and submission module 5 is used to read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. Export control module 6 is used to control the export lock status of the equipment according to three types of receipts. When all three types of receipts are complete, the receipt content is consistent, and the version of the updated receipt record before the update is consistent with the version submitted as the baseline, the equipment export lock status is released based on the export status readback. When any condition is not met, the equipment export lock status is maintained and the transfer voucher is frozen once.

[0136] The above embodiments are used to illustrate the technical solutions of this application. Equivalent substitutions, improvements, or combinations made by those skilled in the art based on the disclosure of this application shall still fall within the protection scope defined by this application.

Claims

1. A method for preventing errors in the flow of compressor parts based on identifier chain verification, characterized in that, include: Step S1: Read the compressor part identifier and the transfer tool identifier at the loading position and the work station entrance, form a co-occupancy record according to the continuous reading interval and the tool occupancy status, and write it into the role grid according to the reading role and role boundary version; Step S2: Verify the identifier chain based on the role grid, the tail segment of the identifier chain, and the process route version to form a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserve one flow credential. Step S3: Read the compressor part identifier, fixture identifier, and machining program identifier at the near end of the equipment, match the reading results with the successor candidate, generate a control session number, and associate it with a primary equipment license and a primary transfer credential; Step S4: Start processing with a single equipment license, and bind the equipment start event, equipment completion event, and in-machine detection event to the control session number, processing cycle identifier, and credential sequence number to form a process execution witness; Step S5: Read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. Step S6: Control the device's exit lock status according to the three types of receipts. When all three types of receipts are complete, the receipt content is consistent, and the version of the updated receipt record before the update is consistent with the version submitted as the baseline, release the device's exit lock status based on the exit status readback. If any condition is not met, maintain the device's exit lock status and freeze the transfer voucher once.

2. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 1, characterized in that, Step S2 specifically includes: The co-occupancy records in the role grid are arranged according to the read role and the role boundary version to form the role sequence corresponding to the same compressor part identifier, and then compared with the end role of the record at the end of the identifier chain. Verify the validity of the role boundary version and the process route version, obtain the next role of the end role according to the valid process route version, match the next role with the first valid role in the role sequence, and filter out co-occupancy records with invalid versions or non-adjacent roles from the role sequence. Compare the compressor part identifier and transfer device identifier in the retained record with the tail segment of the identifier chain, and find the overlapping interval of the continuous reading interval; Based on the comparison results of the two identifiers, the vehicle occupancy status, and the overlapping area, a succession candidate containing the expected reading role and credential succession sequence number is formed; Query the reserved workstations for a single transfer voucher, bind the single transfer voucher that has not been registered for a reserved workstation to the successor candidate, and write the single transfer voucher that has been registered for other workstations into the cross-workstation conflict record.

3. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 2, characterized in that, Step S3 specifically includes: According to the continuous reading interval of the near end of the equipment, the compressor part identifier, fixture identifier and machining program identifier are grouped into the same near end window, and the near end window is closed according to the clamping in place signal; The system continuously matches the read results appearing in the near window with the succession candidates one by one, and verifies the succession sequence number of the voucher. When the compressor part identifier is replaced, the reservation of a transfer voucher is cancelled. When the fixture identifier or machining program identifier does not match, the transfer voucher is frozen. After the reading result matches the continuation candidate, the first-pass credential is changed from the reserved state to the occupied state, a control session number is generated, and the continuation candidate, the near window, and the credential continuation sequence number are associated with the control session number. A device license is generated based on the control session number and written to the device controller. When the same transfer credential corresponds to another control session number or the read result changes, the unexecuted device license is revoked, and the occupancy status of the transfer credential is maintained.

4. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 3, characterized in that, Step S4 specifically includes: Receive the processing cycle identifier returned by the device controller and associate the processing cycle identifier with the control session number and credential sequence number; Obtain equipment start-up and equipment completion events from the equipment controller, obtain in-machine detection events from the in-machine detection interface, and write the three types of events into the execution window corresponding to the processing cycle identifier; Verify the control session number, processing cycle identifier, and credential continuation number of the three types of event records. Form an event sequence for the three types of events that are consistent in all three aspects according to the collection order, and write any event that is inconsistent in any aspect into the conflict event set. The process execution witness is formed based on the succession verification results of the event sequence. When an event is missing or the succession relationship is not established, the corresponding execution window is closed, and the occupancy status of the transfer voucher is retained.

5. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 4, characterized in that, The steps for writing the three types of events into the execution window corresponding to the processing loop identifier specifically include: The execution window is opened based on the equipment start event and stopped based on the equipment completion event. The execution window receives new on-machine detection events and associates the start and end times of the execution window with the processing cycle identifier. The on-machine detection events in the execution window are arranged according to the collection time. The control session number and credential sequence number of the on-machine detection event record are compared. On-machine detection events that do not match are written into the conflict event set. Based on the acquisition time of the equipment completion event, identify the adjacent detection confirmation event from the in-machine detection events that have never been written into the conflict event set, and compare the processing program identifier of the detection confirmation event with the processing program identifier of the primary equipment permission record; When the processing program identifiers of the detection and confirmation events are consistent, the equipment start event, equipment completion event, and detection and confirmation event are identified as the valid event set of the execution window corresponding to the processing cycle identifier. When the detection and confirmation event is missing or the processing program identifiers are inconsistent, the execution window is marked as an event gap.

6. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 1, characterized in that, Step S5 specifically includes: Based on the process execution witness, a submission batch number and a submission sequence are generated. The write sequence number, the verification sequence number, and the update sequence number are obtained from the submission sequence. The chain tail version that identifies the chain tail segment is recorded as the submission base version. The control session number and processing cycle identifier are assigned to the session item, the credential continuation sequence number and continuation candidate are assigned to the continuation item, and the session item and continuation item are combined to form the submission payload. Write the batch number into the process confirmation segment and obtain a write receipt that records the batch number, write sequence number, and submitted payload. Once a transfer voucher is cancelled based on the submitted batch number, a cancellation receipt is obtained after the voucher sequence number matches, recording the submitted batch number, cancellation sequence number, and submitted load. After the tail version is consistent with the submission baseline version, update the tail segment to the process confirmation segment and obtain the update receipts for the submission batch number, update sequence number, submission payload, tail version before update and tail version after update. Verify the write receipt, cancellation receipt, and update receipt, and generate a submission status based on the verification results.

7. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 6, characterized in that, The steps for verifying the write receipt, cancel receipt, and update receipt include: The three types of receipts are grouped into a receipt set. Receipts that are not from the current submission batch number are filtered out by receipt type and corresponding serial number. Compare the session items and continuation items in the receipt set, and record any inconsistency as a load conflict; Compare the order of writing sequence number, canceling sequence number, and updating sequence number, and record duplicate or reversed sequence numbers as operation sequence number conflicts; Compare the sequential number of the write-off receipt and the document for witnessing the execution of the process. Record the smaller former as a historical replay and the larger former as a sequential conflict. Compare the pre-update tail version of the update receipt with the submission baseline version. Record inconsistencies in the pre-update tail version, operation sequence number conflicts, or historical replays as timing conflicts, and record continuation conflicts as payload conflicts. When there are no receipt gaps, load conflicts, or timing conflicts in the receipt set, the receipt set is recorded as a complete submission status; When any issue exists, the receipt set is marked as an incomplete submission status, and the corresponding issue is written to the submission status.

8. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 7, characterized in that, Step S6 specifically includes: Associate the equipment's export lock status with the submission batch number to generate an export lock serial number, and write the complete or incomplete submission status into the export control record. Based on the incomplete submission status, the equipment exit lock status is maintained and the transfer voucher is frozen once. Subsequent equipment licenses corresponding to the same compressor part identifier are refused, and the exit lock serial number is written into the rejection record. Based on the complete submission status, compare the process confirmation identifier segment of the updated receipt record with the process confirmation identifier segment of the identifier chain tail record. If they match, generate an unlocking instruction containing the exit lock sequence number and control session number. Write the unlock command to the device controller, receive the exit status readback containing the exit lock sequence number and control session number, compare the two items in the exit status readback with the unlock command, when the two items match, write the unlock confirmation and release the device exit lock status, when either item does not match, write the lock confirmation and keep the device exit lock status.

9. The compressor parts transfer error prevention method based on identifier chain verification as described in claim 8, characterized in that, The steps for maintaining the device's exit lock status and freezing the transfer credentials once based on the incomplete submission status specifically include: Based on the submitted batch number, obtain the receipt set and process execution witness, and compare the control session number, processing cycle identifier, and document continuation number in the receipt set with the corresponding content of the process execution witness; Obtain the processing cycle identifier and voucher sequence number from the equipment controller, compare the two items with the corresponding items in the process execution witness, and record any inconsistency as an execution conflict; When all items are consistent and only the receipt is missing, the corresponding problem is recorded as a recoverable gap. The equipment exit is kept locked and the transfer voucher is frozen once. The missing operation is performed using the submitted batch number, and the new receipt is written into the receipt set. When load conflicts, timing conflicts, or execution conflicts exist, the corresponding issues will be recorded as isolation conflicts, the missing operations will be refused to be executed, the equipment exit will be kept locked and the transfer voucher will be frozen, and the compressor part identification and submission batch number will be written into the isolation record.

10. A compressor parts transfer error prevention system based on identifier chain verification, characterized in that, include: The co-occupancy write module is used to read compressor part identifiers and transfer fixture identifiers at loading positions and workstation entrances, form co-occupancy records according to continuous reading intervals and fixture occupancy status, and write them to the role grid according to the reading role and role boundary version. The identifier verification module is used to verify the identifier chain based on the role grid, the tail of the identifier chain, and the process route version, forming a succession candidate containing the expected reading role and the credential succession sequence number, and exclusively reserving one flow credential. The session authorization module is used to read compressor part identifiers, fixture identifiers, and machining program identifiers at the near end of the equipment, match the reading results with success candidates, generate a control session number, and associate a device authorization with a transfer credential. The witness generation module is used to start processing with a single equipment license, and bind the equipment start event, equipment completion event and in-machine detection event to the control session number, processing cycle identifier and credential sequence number to form a process execution witness; The confirmation and submission module is used to read the tail version of the chain based on the process execution witness as the submission baseline version, write the process confirmation identifier segment according to the submission batch number, cancel the first-time circulation voucher and update the identifier tail segment, and obtain the write receipt, cancellation receipt and update receipt. The export control module is used to control the export lock status of the equipment according to three types of receipts. When all three types of receipts are complete, the content of the receipts is consistent, and the version of the chain tail before the update of the updated receipt record is consistent with the version submitted as the baseline, the export lock status of the equipment is released based on the export status readback. When any condition is not met, the export lock status of the equipment is maintained and the transfer voucher is frozen once.

Images