A method, device, equipment and medium for link construction of a black light laboratory sample

Abstract

The application discloses a black lamp laboratory sample link construction method, device, equipment and medium, and belongs to the field of link construction. The method is: obtaining sample processing data of a black lamp laboratory sample, and constructing a sample event time sequence chain by using the sample processing data; respectively calculating a credibility score corresponding to each sample event in the sample event time sequence chain and a consistency score corresponding to two adjacent sample events, to obtain the credibility score value and the consistency score value; filtering events of the sample event time sequence chain according to the credibility score value and the consistency score value, and constructing a global optimal link of the sample event; determining that the global optimal link has a link breakpoint according to a transfer relationship of two adjacent sample events in the global optimal link, and then performing a completion treatment on the global optimal link to obtain an enhanced link, and performing time restoration on the enhanced link to obtain a sample processing link. The application can complete the link, and ensures the precision and accuracy of the link.

Description

Technical Field

[0001] This invention relates to the technical field of link construction, and in particular to a method, apparatus, equipment and medium for constructing links for samples in a dark laboratory. Background Technology

[0002] In building engineering testing laboratories, samples such as concrete test blocks, reinforcing steel raw materials, welded reinforcing steel specimens, and mechanically connected reinforcing steel specimens undergo multiple processes including sample receipt and registration, temporary storage, sample sorting, transportation, curing, testing, and report generation. The transparency and traceability of this process directly affect the reliability and credibility of the test results. Therefore, it is necessary to construct a data link for the entire process to enable technicians to trace samples.

[0003] With the gradual promotion of the "lights-out lab" model, testing processes increasingly rely on automated equipment and unattended operation. One common approach for samples in lights-out labs is to define a pre-set processing flow for the samples, identify the automated equipment corresponding to each step of the flow, acquire processing data from each device according to the time sequence of the processing flow, and construct a complete processing chain based on this data.

[0004] However, the above method has the following technical problems: during the sample processing in the dark laboratory, various abnormalities or special situations may occur, requiring manual intervention or transfer to other processes for abnormal handling. Based on the pre-set processing process, data is extracted from the equipment to build a data link, but there may be data loss, resulting in deviations between the link and the actual situation. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for constructing links for samples in a dark laboratory, which can solve one or more technical problems existing in the prior art.

[0006] A first aspect of this invention provides a method for constructing a link in a light-down laboratory sample, the method comprising: Acquire sample processing data of samples from a dark laboratory, and construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and recorded business processing data; The credibility score and the consistency score of each sample event in the sample event time series chain are calculated respectively to obtain the credibility score value and the consistency score value. The events in the sample event time series chain are filtered based on the credibility score and the consistency score, and a globally optimal link for the sample events is constructed. Determine whether there is a link breakpoint in the global optimal link based on the transfer relationship between two adjacent sample events in the global optimal link; If it is determined that there is a link breakpoint in the globally optimal link, then the globally optimal link is completed to obtain an enhanced link, and the enhanced link is time-restored to obtain a sample processing link.

[0007] In conjunction with the first aspect, in one implementation, the step of calculating the confidence score corresponding to each sample event in the sample event time series chain and the consistency score corresponding to two adjacent sample events to obtain the confidence score value and the consistency score value includes: Based on a preset set of fixed spatial anchor points and a preset set of transport path segments, each sample event in the sample event time series chain is converted into a candidate spatial anchor point and combined into a candidate spatial anchor point set. The candidate spatial anchor point set includes multiple candidate spatial anchor points, and each candidate spatial anchor point corresponds to a sample event. Determine the credibility parameter corresponding to each candidate spatial anchor point, and calculate the credibility score using the credibility parameters corresponding to two adjacent candidate spatial anchor points. The credibility parameter includes: comprehensive rule-based score, event source reliability factor, and global conflict penalty item. Determine the consistency parameters corresponding to each candidate spatial anchor point, and calculate the consistency score using the consistency parameters corresponding to two adjacent candidate spatial anchor points. The consistency parameters include: timing constraints, process constraints, path constraints, and conflict penalty terms.

[0008] In conjunction with the first aspect, in one implementation, the step of filtering the events in the sample event time series chain based on the credibility score and the consistency score, and constructing the globally optimal link of the sample events, includes: Based on the credibility score, candidate spatial anchors that meet the preset credibility threshold are selected from the candidate spatial anchor set corresponding to the sample event time series chain as filter anchors. Based on the consistency score, consistent anchors and inconsistent anchors are selected from the filter anchors. The consistent anchors are filter anchors whose consistency scores meet a preset consistency threshold, and the inconsistent anchors are filter anchors whose consistency scores do not meet the preset consistency threshold but meet timing constraints, process constraints, and path constraints. A hierarchical extended state diagram is constructed using the consistent anchor points and the inconsistent anchor points, and the hierarchical extended state diagram is solved to obtain the globally optimal link.

[0009] In conjunction with the first aspect, in one implementation, determining whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events in the globally optimal link includes: The processing information for identifying two adjacent sample events within the globally optimal link includes: event time, stage semantics, candidate spatial anchor point, sample type, process flow template, and auxiliary record information. Based on the processing information, determine whether there is a direct transfer relationship between two adjacent sample events; If it is determined that there is a direct transfer relationship between every two adjacent sample events, then it is determined that there is no link breakpoint in the global optimal link; If it is determined that there is no direct transfer relationship between any two adjacent sample events, then it is determined that there is a link breakpoint in the globally optimal link.

[0010] In conjunction with the first aspect, in one implementation, determining whether there is a direct transfer relationship between two adjacent sample events based on the processing information includes: Each of the following criteria is determined: whether the sum of the times of the two events covers the preset process time; whether the semantics of the two stages are within the process flow template corresponding to the sample type; whether there is a communication laboratory path between the two candidate spatial anchors; and whether there is an information conflict between the two auxiliary record information and the personnel record information of the technicians. If the sum of the times of the two events covers the preset process time, the semantics of the two stages are within the process flow template corresponding to the sample type, the two candidate spatial anchors have a communicating laboratory path, and the two auxiliary record information and the personnel record information of the technicians do not conflict, then it is determined that there is a direct transfer relationship between the two adjacent sample events. If the sum of the times of the two events does not cover the preset process time, the semantics of the two stages are not within the process flow template corresponding to the sample type, the two candidate spatial anchors do not have a communication laboratory path, or the two auxiliary record information conflicts with the personnel record information of the technicians, then it is determined that there is no direct transfer relationship between the two adjacent sample events.

[0011] In conjunction with the first aspect, in one implementation, the step of performing time restoration on the globally optimal link completion process to obtain an enhanced link, and performing time restoration on the enhanced link to obtain a sample processing link, includes: Determine the link breakpoint of the globally optimal link and construct a set of virtual segments using the link breakpoint, wherein the link breakpoint is an adjacent sample event pair whose consistency score does not meet the preset consistency threshold and does not meet the timing constraints, process constraints and path constraints; The virtual segment set is used to filter and complete segments, wherein the completed segments are virtual segments that meet the preset reasonable score in terms of completion time reachability, process rationality and path connectivity, and the completion rationality score is satisfied. Each virtual segment is constructed based on the link breakpoint. The completed segment is inserted into the globally optimal link to obtain an enhanced link, and the time slice corresponding to the sample processing data is inserted into the enhanced link to obtain a sample processing link.

[0012] In conjunction with the first aspect, in one implementation, after the steps of completing the globally optimal link to obtain the enhanced link and performing time restoration on the enhanced link to obtain the sample processing link, the method further includes: Extract the segment information corresponding to the sample event in the time segment from the sample processing link, wherein the segment information includes: spatial anchor point, path segment, segment type, start and end time, current position, stage to which it belongs and anomaly description information; The segment information is visualized, which includes visualizing the segment information in chronological order, locating real or virtual segments in the visualization sequence, and adjusting the animation and style according to the segment type.

[0013] A second aspect of the present invention provides a link construction apparatus for samples in a dark laboratory, the apparatus comprising: The acquisition module is used to acquire sample processing data of samples from a dark laboratory and to construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and record business processing data. The calculation module is used to calculate the confidence score corresponding to each sample event in the sample event time series chain and the consistency score corresponding to two adjacent sample events, so as to obtain the confidence score value and the consistency score value. The filtering module is used to filter the events in the sample event time series chain based on the credibility score and the consistency score, and to construct the globally optimal link of the sample events. The determination module is used to determine whether there is a link breakpoint in the global optimal link based on the transfer relationship between two adjacent sample events in the global optimal link; The construction module is used to complete the globally optimal link to obtain an enhanced link if it is determined that there is a link breakpoint in the globally optimal link, and to perform time restoration on the enhanced link to obtain a sample processing link.

[0014] Compared to existing technologies, the present invention provides a method, apparatus, device, and medium for constructing a link for samples in a dark laboratory. The advantages are as follows: The present invention can acquire sample processing data of samples in a dark laboratory and construct a time series chain of sample events using this data; calculate the confidence score for each sample event in the time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and consistency score value; filter the events in the time series chain based on the confidence score value and consistency score value and construct a globally optimal link for the sample events; determine whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events; if a link breakpoint is determined to exist in the globally optimal link, complete the globally optimal link to obtain an enhanced link, and perform time restoration on the enhanced link to obtain the sample processing link. When constructing the link, the present invention can determine whether there is a link breakpoint in the globally optimal link based on the transition relationship between two sample events, and complete the link when a breakpoint exists, thus avoiding missing link data and ensuring the accuracy and precision of the link. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a method for constructing a link for a sample in a dark laboratory according to an embodiment of the present invention. Figure 2 This is an operation flowchart of a method for constructing a link for a sample in a dark laboratory according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a link construction device for a dark laboratory sample provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] In building engineering testing laboratories, samples such as concrete test blocks, reinforcing steel raw materials, welded reinforcing steel specimens, and mechanically connected reinforcing steel specimens undergo multiple processes including sample receipt and registration, temporary storage, sample sorting, transportation, curing, testing, and report generation. The transparency and traceability of this process directly affect the reliability and credibility of the test results. Therefore, it is necessary to construct a data link for the entire process to enable technicians to trace samples.

[0018] With the gradual promotion of the "lights-out lab" model, testing processes increasingly rely on automated equipment and unattended operation. One common approach for samples in lights-out labs is to define a pre-set processing flow for the samples, identify the automated equipment corresponding to each step of the flow, acquire processing data from each device according to the time sequence of the processing flow, and construct a complete processing chain based on this data.

[0019] However, the above method has the following technical problems: during the sample processing in the dark laboratory, various abnormalities or special situations may occur, requiring manual intervention or transfer to other processes for abnormal handling. Based on the pre-set processing process, data is extracted from the equipment to build a data link, but there may be data loss, resulting in deviations between the link and the actual situation.

[0020] To address the aforementioned issues, the following specific embodiments will provide a detailed description and explanation of a method, apparatus, device, and medium for constructing a link for a dark laboratory sample, as provided in this application.

[0021] To address the technical issues of missing data links and low accuracy in existing technologies, referencing Figure 1 The diagram shows a flowchart of a method for constructing a link for a sample in a dark laboratory according to an embodiment of the present invention.

[0022] As an example, the link construction method for the sample in the dark laboratory may include: S11. Obtain sample processing data of samples from a dark laboratory, and construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and recorded business processing data.

[0023] In one embodiment, sample processing data of samples from a dark laboratory can be obtained. The sample processing data includes sample characteristic data, which may be the master sample data of the target sample in the laboratory. Specifically, it may include identifiers such as sample number, sample type, order number, submitting unit, submitter, current status, and affiliated laboratory.

[0024] The sample processing data also includes recorded business processing data, which can be the original business records of the target sample during laboratory operation, including laboratory management system records, equipment records, AGV records, business records in the manual operating system, etc.

[0025] Next, a time series chain of sample events can be constructed using the sample processing data. The raw records from different business subsystems are standardized to achieve a unified representation of the entire sample event process, and this is defined as a sample event according to the following standard, as shown in the following formula: ; in: Indicates the sample event identifier; Indicates the time of occurrence of the sample event; Indicates the sample event category; This represents the collection of business fields for sample events; Indicates stage semantics; This represents the reliability factor of the sample event source.

[0026] The sample event identifier uniquely identifies a standard sample event; the sample event occurrence time characterizes the temporal position of the sample event in the sample process; the sample event category distinguishes different business types such as sample receipt, temporary storage, maintenance, transfer, machine loading, test completion, and anomaly handling; the sample event business field set carries the associated equipment code, area code, workstation code, task number, operator identifier, sample status value, and other original business fields; the stage semantics unifies and standardizes original records from different sources but with similar business meanings into preset test process stages; and the sample event source reliability factor indicates the credibility of the sample event source. The source reliability of sample events automatically collected by equipment is higher than that of manually recorded events, and the source reliability of sample events directly generated by the main business system is higher than that of events forwarded through intermediate interfaces.

[0027] In one embodiment, standardization can be performed according to the following standard sample event rules, specifically, the standard sample event rules include the following processing steps: Extract the sample number, order number, task number, equipment number, area number, workstation number, status code, operation time, operator identification, and other basic fields related to sample flow from the original records.

[0028] Map sample numbers, order numbers, task numbers, etc. from different systems to a unified associated primary key to establish a cross-system relationship between records.

[0029] Convert time fields from different systems to the same time base and handle differences in time at the second, millisecond, or supplementary recording levels.

[0030] By mapping status codes, event codes, device codes, and region codes from different systems to a unified business dictionary, the coding differences between different subsystems are eliminated.

[0031] The different status expressions such as "sample received", "stored", "awaiting maintenance", "under testing", and "abnormal" are organized into a preset stage semantic set.

[0032] Identify the source system type of the original records and record the source system identifier. Assign a source reliability factor to each sample event based on the source type of the sample event.

[0033] Standard sample event objects are output according to a unified field structure to serve as the basic input for subsequent candidate spatial anchor point generation, link solving, virtual segment completion, and time slice state restoration.

[0034] Original records from different systems are mapped to standard sample event objects according to their data meanings. Laboratory management system records are preferentially mapped to the fields of sample number, order number, task number, event category, stage semantics, and business status; equipment records are preferentially mapped to the fields of equipment code, workstation code, test start time, test completion time, equipment status, and anomaly identifier; AGV records are preferentially mapped to the fields of transfer task number, starting area, ending area, path segment, task start time, task completion time, and task status; and manual operation records are preferentially mapped to the fields of operator identifier, manual handover time, manual handling start point, manual handling end point, review instructions, and anomaly instructions. After completing the above mapping, the original records can be organized according to a unified event category dictionary, a unified stage semantic dictionary, and a unified spatial coding dictionary to generate a set of standard sample events, and stage semantic annotations can be completed for each standard sample event. The annotated stages are sample receipt stage, temporary storage stage, preservation stage, transfer stage, testing stage, and anomaly handling stage, etc. All standard events are arranged in ascending order of time to form a sample event time sequence chain.

[0035] It should be noted that constructing the sample event time series chain is the ordered input foundation for subsequent processing, providing contextual constraints for candidate spatial anchor point generation. It is used to identify the legal location range of the previous, next, and current stages, thereby narrowing down the candidate location set. It is not directly used to uniquely map discrete records to a single spatial location. As the vertical hierarchical basis for the hierarchical extended state diagram construction, it is used to determine the order of each standard sample event in the link recovery solution. When adjacent events cannot form a direct transition relationship that satisfies the constraints, the link breakpoint can be identified based on this time series chain, triggering the subsequent virtual segment completion process. During time slice state restoration, the sample event time series chain is used to determine which real event segment or virtual segment any given time should fall into.

[0036] Specifically, the sample event time series chain can be represented by the following equation: .

[0037] S12. Calculate the confidence score for each sample event in the sample event time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and the consistency score value.

[0038] In one embodiment, the sample event time series chain includes multiple sample events, and the multiple sample events are sorted. The credibility score corresponding to each sample event in the sample event time series chain can be calculated sequentially according to the order of the multiple sample events to obtain the credibility score value. Then, the consistency score corresponding to two adjacent sample events can be calculated to obtain the consistency score value.

[0039] In an optional embodiment, the step of calculating the confidence score for each sample event in the sample event time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and the consistency score value may include the following sub-steps: S121. Based on a preset set of fixed spatial anchor points and a preset set of transport path segments, each sample event in the sample event time series chain is converted into a candidate spatial anchor point and combined into a candidate spatial anchor point set, wherein the candidate spatial anchor point set includes multiple candidate spatial anchor points, and each candidate spatial anchor point corresponds to a sample event.

[0040] S122. Determine the credibility parameter corresponding to each candidate spatial anchor point, and calculate the credibility score using the credibility parameters corresponding to two adjacent candidate spatial anchor points. The credibility parameter includes: comprehensive rule-based score, event source reliability factor and global conflict penalty item.

[0041] S123. Determine the consistency parameters corresponding to each candidate spatial anchor point, and calculate the consistency score using the consistency parameters corresponding to two adjacent candidate spatial anchor points. The consistency parameters include: timing constraints, process constraints, path constraints, and conflict penalty terms.

[0042] In one embodiment, a set of fixed spatial anchor point units is pre-configured based on event categories, equipment identifiers, area identifiers, workstation identifiers, etc., in the laboratory. and the set of preset transfer path segments For any sample event Instead of directly and uniquely mapping to a single spatial location, each sample event in the sample event time series chain is transformed into a candidate spatial anchor point and combined into a set of candidate spatial anchor points. Specifically, the set of candidate spatial anchor points can be represented by the following formula: ; right An event credibility score is calculated for each candidate spatial anchor point v. Specifically, a credibility parameter corresponding to each candidate spatial anchor point can be determined, and the credibility score value is calculated using the credibility parameters corresponding to two adjacent candidate spatial anchor points. The credibility parameters include: a comprehensive rule-based score, an event source reliability factor, and a global conflict penalty term.

[0043] In one operational mode, the event credibility score is calculated using a gated rule-based scoring method. First, a global hard constraint screening is performed on candidate spatial anchor points v. If a candidate spatial anchor point v exhibits absolutely unreasonable circumstances such as sample type prohibition, incorrect equipment attribution, inaccessible region, conflicting process stages, unreachable time, or irresolvable conflict with a confirmed event, then the candidate spatial anchor point is excluded from the candidate set, and the event credibility score is no longer calculated. If the candidate spatial anchor point v passes the global hard constraint screening, then the event credibility score is calculated.

[0044] In an optional embodiment, candidate shrinking can be performed first based on the event field, and then the confidence score of the shrunken candidate spatial anchor points can be applied.

[0045] The candidate shrinking rules include at least the following: when an event contains a clear equipment code, the equipment bit corresponding to the equipment code and its adjacent buffer bit are preferentially included in the candidate set; when an event contains a region code but not a equipment code, the workstation, rack position or waiting position in the region that semantically matches the current stage is included in the candidate set; when an event belongs to AGV transfer or manual transfer, the corresponding starting position, ending position and allowed path segment are included in the candidate set; when an event only contains status information and has no clear location field, the candidate set is generated by combining the legal process location range of the previous stage and the next stage of the sample.

[0046] Candidate spatial anchor point generation does not immediately and uniquely assign a single discrete record to a specific spatial location. Instead, it avoids misjudgments due to incomplete event information caused by a single mapping. In a dark laboratory setting, a single event often contains only partial fields from equipment number, area number, task number, or stage status. These fields typically only limit the sample location to a range of possible positions, rather than reliably and uniquely pointing to a specific workstation, rack, or equipment position. For example, multiple buffer positions, inspection positions, or equipment buffer positions may exist simultaneously within the same area; the same status event may also correspond to multiple legitimate process positions. Directly performing a single mapping on such events can easily lead to incorrect sample positioning and further affect the subsequent link recovery results.

[0047] Specifically, the credibility score can be represented by the following formula: ; in, This represents the event source reliability factor for the i-th standard sample event. This represents a global conflict penalty term, used to reduce the score of candidate spatial anchors that pass hard constraints but still have relative conflicts. For comprehensive rule-based scoring.

[0048] Specifically, the comprehensive regularized score of the candidate spatial anchor points can be expressed as follows: ; In the above formula, The number of consensus functions actually involved in the scoring can be calculated as follows: ; , , , These four enable factor parameters indicate whether there is enough information to support this evaluation, with 1 for applicable and 0 for not applicable.

[0049] When all four enable factors are 0, candidate spatial anchor point scores are not calculated. When a standard sample event lacks sufficient fields to calculate a candidate spatial anchor point score, the range of possible legal process locations corresponding to the event is determined based on the preceding and following adjacent events, sample type, and corresponding process flow template. Contextual candidate spatial anchor points are then generated from this range. If a valid candidate spatial anchor point cannot be generated based on the context, it is treated as a link breakpoint in subsequent adjacent event transition judgments, and the virtual segment completion process is used to further determine whether a completion process needs to be generated.

[0050] This represents the type-based scoring value determined by a preset matching rule between the event category and the candidate spatial anchor point type. For example, a maintenance event receives a high score when it corresponds to a maintenance position, a medium score when it corresponds to a maintenance area cache position, a low score when it corresponds to a testing equipment position, and 0 when it corresponds to an irrelevant area such as the report printing area; a machine testing event receives a high score when it corresponds to a testing equipment position, a medium score when it corresponds to a front cache position of the equipment, and a low score when it corresponds to a maintenance position.

[0051] This represents the standardized score for the equipment determined based on the equipment code, workstation code, and task record. A higher score is given when the equipment code is exactly the same as the equipment bound to the candidate spatial anchor; a lower score is given when the equipment code and the equipment bound to the candidate spatial anchor are aliases or equivalent codes; a lower score is given when the equipment code and the equipment bound to the candidate spatial anchor belong to the same equipment group; a higher score is given when the workstation code and the workstation bound to the candidate spatial anchor are the same; a lower score is given when the workstation code and the workstation bound to the candidate spatial anchor are adjacent workstations. If only a task record supports the candidate spatial anchor, the score is determined based on task time overlap and task start-end distance; a score of 0 is given if no equipment, workstation, or task record supports it.

[0052] This represents the region regularization score determined based on region coding, region hierarchy, functional area relationships, and access rules. A high score is awarded when the event region code is completely identical to the region to which the candidate spatial anchor point belongs. A medium score is awarded when the two are parent and child regions or belong to the same functional area. A medium score is awarded when the two are adjacent process functional areas. A low score is awarded when the path is reachable but the region semantics are weak.

[0053] This represents the stage-based rule-based score determined according to preset matching rules between stage semantics and process roles. For stage-based rule-based scores, a high score is awarded when the stage semantics and the process role of the candidate spatial anchor are completely consistent. A medium score is awarded when the candidate spatial anchor belongs to a permitted preceding / following role, buffer role, or transition role within that stage. A low score is awarded when the candidate spatial anchor, although not a direct process role within that stage, has a weak process association with that stage and does not constitute a conflict, based on task records or contextual events. For example, a high score is awarded when the temporary storage stage corresponds to a temporary storage rack or buffer position. A medium score is awarded when the detection stage corresponds to a buffer position before equipment or a buffer position awaiting inspection. A low score is awarded when the maintenance stage corresponds to a maintenance completion transfer position or a maintenance area exit buffer position.

[0054] For adjacent sample events and The corresponding candidate spatial anchor points are respectively and The consistency parameters corresponding to each candidate spatial anchor point are determined, and a consistency score is calculated using the consistency parameters corresponding to two adjacent candidate spatial anchor points. The consistency parameters include: timing constraints, process constraints, path constraints, and conflict penalty terms. Specifically, the consistency score can be a calculated value corresponding to a transfer consistency score. The consistency score can be calculated as follows: ; in, Indicates the time difference between adjacent events; This represents a timing constraint function used to determine whether a sample can move from one position to the next within a given time period. This represents a process constraint function used to determine the sample type. Corresponding process template Next, whether adjacent transitions are allowed in the stages to which the two events belong; This represents a path constraint function used to determine the path in the laboratory path diagram. Is there a permitted transit route between the two anchor points? This indicates a conflict penalty item, used to punish unreasonable situations such as illegal jumps, time inversions, and cross-regional jumps.

[0055] This indicates the weighting of the transfer score. The temporal constraint function does not only determine the order of events, but also considers the shortest reachable path length between the previous and next anchor points, the permitted transfer method, and the shortest dwell or transfer time for this type of sample in the corresponding section to determine whether reachability exists within the time difference between adjacent events. The process constraint function is used to determine whether direct connection between two stages is allowed based on a preset sample process template. For example, after sample receipt, it can enter the temporary storage or curing stage, but it cannot directly enter the report generation stage without a necessary intermediate stage. The path constraint function is used to determine whether a permissible path exists between two anchor points based on the laboratory path map, regional access rules, and equipment layout. The conflict penalty term is used to penalize abnormal transfers such as time reversal, reverse process jumps, cross-regional skip-level jumps, and forced connections without a path.

[0056] S13. Filter the events in the sample event time series chain according to the credibility score and the consistency score, and construct the globally optimal link of the sample events.

[0057] In one embodiment, events in the sample event time series chain can be filtered based on the confidence score and the consistency score, and then the global optimal link of sample events can be constructed based on the remaining sample information after filtering.

[0058] In one operation mode, sample events correspond to candidate spatial anchors. Candidate spatial anchors in the candidate spatial anchor set can be filtered and selected based on the confidence score and consistency score. Then, the globally optimal link of sample events is constructed using the selected candidate spatial anchors.

[0059] In one embodiment, the step of filtering the events in the sample event time series chain based on the confidence score and the consistency score, and constructing the globally optimal link of the sample events, may include the following sub-steps: S131. Based on the credibility score, select candidate spatial anchors that meet the preset credibility threshold from the candidate spatial anchor set corresponding to the sample event time series chain as filter anchors.

[0060] S132. Based on the consistency score, filter consistent anchors and inconsistent anchors from the filter anchors, wherein the consistent anchors are filter anchors whose consistency score satisfies a preset consistency threshold, and the inconsistent anchors are filter anchors whose consistency score does not satisfy the preset consistency threshold but satisfy timing constraints, process constraints, and path constraints.

[0061] S133. Construct a hierarchical extended state diagram using the consistent anchor points and inconsistent anchor points, and solve the hierarchical extended state diagram to obtain the globally optimal link.

[0062] In one embodiment, candidate spatial anchors that meet a preset confidence threshold can be selected as filter anchors from the set of candidate spatial anchors corresponding to the sample event time series chain based on the confidence score.

[0063] Specifically, candidate spatial anchors with credibility scores greater than a preset credibility threshold can be selected as filter anchors. Based on the aforementioned credibility scores, a set of candidate spatial anchors with scores above the threshold are retained for each event, which are then used for subsequent link recovery.

[0064] Next, consistent anchors and inconsistent anchors can be filtered from the filter anchors based on the consistency score value. The consistent anchors are filter anchors whose consistency score value meets the preset consistency threshold, and the inconsistent anchors are filter anchors whose consistency score value does not meet the preset consistency threshold but meets the timing constraints, process constraints, and path constraints.

[0065] Then, a hierarchical extended state diagram is constructed using consistent anchor points and inconsistent anchor points, and the hierarchical extended state diagram is solved to obtain the globally optimal link.

[0066] If there is at least one pair of candidate spatial anchor points between two adjacent events, which simultaneously satisfy timing constraints, process constraints, and path constraints, and whose consistency score is higher than the preset transfer threshold, then it is determined that there is a direct transfer relationship between the adjacent events. If there is no pair of candidate spatial anchor points between two adjacent events that satisfies the above conditions, or if there is a pair of candidate spatial anchor points but whose consistency scores are all lower than the preset transfer threshold, then it is determined that there is no direct transfer relationship between the adjacent events, and the location is marked as a link breakpoint to trigger the subsequent virtual segment completion process.

[0067] Specifically, under the above constraints, constructing the extended state diagram and finding the globally optimal link can be done as follows: ; in, Represents the set of all feasible links. This indicates that a missing penalty item needs to be filled in. This indicates a penalty for inconsistencies. and The penalty weight is indicated by M(Π). The missing completion penalty term M(Π) is used to measure the cost of the virtual completion process introduced in the sample link due to the missing original record. This cost is related to the number of virtual segments, the duration of the virtual segment, the length of the path crossed by the virtual segment, and the support strength of the auxiliary record.

[0068] The more virtual segments introduced into the sample link, the longer the duration, and the greater the span, the larger the missing completion penalty term M(Π).

[0069] The inconsistency penalty term B(Π) measures the degree of inconsistency between the sample link and the anomaly record, process rules, and business facts. If the recovered sample link does not pass through the anomaly handling section corresponding to the anomaly record, is inconsistent with the manual handover record, or conflicts with the sample process template, the inconsistency penalty term B(Π) increases. For any candidate state node v in the i-th layer, its cumulative optimal score is defined as... The cumulative optimal score is determined using the following recursive method: ; in, This represents the cumulative optimal score for reaching the candidate state node v at level i. Let represent the event confidence score between the i-th standard sample event and the candidate spatial anchor point v. Let u represent the set of candidate spatial anchor points in the (i-1)th layer, and let u represent any candidate state node in the (i-1)th layer. This represents the consistency score of the transition from candidate state node u in the previous layer to candidate state node v in the current layer. The cumulative best score. This represents the optimal score for the stage when reaching candidate state node v in the i-th layer from the first event layer. After calculating the cumulative optimal score of all candidate state nodes recursively layer by layer, the candidate state node with the highest cumulative optimal score is selected as the termination node in the final event layer. ; in, This represents the set of candidate spatial anchor points for the final event layer. This represents the cumulative optimal score of candidate state node v in the final event layer. The cumulative optimal score is calculated for each candidate state node. Simultaneously, record the predecessor candidate state node that maximizes the score, and determine the termination node. Then, by backtracking back according to the predecessor candidate state nodes recorded in each layer, the complete original optimal sample link is obtained.

[0070] In one embodiment, the solution process includes the following steps: Using each event in the sample event time series chain as a vertical level, and the candidate spatial anchor point corresponding to each standard sample event as the candidate state node of that level, a hierarchical extended state diagram is constructed.

[0071] Calculate a node score for each candidate spatial anchor node based on the event credibility scoring function. Also, calculate edge scores for candidate state pairs between adjacent event layers based on the transition consistency scoring function.

[0072] Determine whether a path is transferable by combining timing constraints, process constraints, and path constraints; if it is transferable, establish a transfer edge. If it is clearly not transferable, do not establish a feasible edge; if there is local information missing but the possibility of subsequent completion still needs to be retained, establish a low-confidence transfer edge with conflict penalty.

[0073] Starting from the first event layer, a phased accumulation method using dynamic programming is adopted to calculate the cumulative optimal score reaching each candidate state node.

[0074] In the final event layer, the termination node with the highest cumulative score is selected, and the process is backtracked along the predecessor candidate state nodes recorded in each layer to obtain the complete optimal sample link. The output includes the selected spatial anchor point for each event, the transition relationship between adjacent events, the cumulative link score, and the location of any breakpoints.

[0075] S14. Determine whether there is a link breakpoint in the global optimal link based on the transition relationship between two adjacent sample events in the global optimal link.

[0076] In one embodiment, the transition relationship between two adjacent sample events in the globally optimal link can be determined. Specifically, the transition relationship between two adjacent candidate spatial anchors in the globally optimal link can be determined. Then, based on the transition relationship between the two adjacent candidate spatial anchors in the globally optimal link, it is determined whether there is a breakpoint between the two adjacent candidate spatial anchors. If there is, it is determined that there is a link breakpoint in the globally optimal link; otherwise, it is determined that there is no link breakpoint in the globally optimal link.

[0077] In one embodiment, determining whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events in the globally optimal link may include the following sub-steps: S141. Identify the processing information of two adjacent sample events within the globally optimal link, wherein the processing information includes: event time, stage semantics, candidate spatial anchor point, sample type, process flow template, and auxiliary record information.

[0078] S142. Determine whether there is a direct transfer relationship between two adjacent sample events based on the processing information.

[0079] S143. If it is determined that there is a direct transfer relationship between each pair of adjacent sample events, then it is determined that there is no link breakpoint in the global optimal link.

[0080] S144. If it is determined that there is no direct transfer relationship between any two adjacent sample events, then it is determined that there is a link breakpoint in the global optimal link.

[0081] In one embodiment, after constructing the sample event time series chain, a direct connection judgment is performed on any adjacent event pair according to the chronological order of event occurrence. For the preceding and following events, the processing information of both events is read. The processing information may include event time, stage semantics, candidate spatial anchor points, sample type, process flow template, and auxiliary record information, and it is determined whether a credible direct transfer relationship can be formed between the two events.

[0082] If it is determined that there is a direct transfer relationship between every two adjacent sample events, then it is determined that there are no link breaks in the globally optimal link.

[0083] Conversely, if it is determined that there is no direct transfer relationship between any two adjacent sample events, then it is determined that there is a link breakpoint in the globally optimal link.

[0084] As an example, determining whether there is a direct transfer relationship between two adjacent sample events based on the processing information may include the following sub-steps: S1421. Determine whether the sum of the times of the two events covers the preset process time, whether the semantics of the two stages are within the process flow template corresponding to the sample type, whether there is a communication laboratory path between the two candidate spatial anchors, and whether there is an information conflict between the two auxiliary record information and the personnel record information of the technicians.

[0085] S1422. If the sum of the times of the two events covers the preset process time, the semantics of the two stages are within the process flow template corresponding to the sample type, the two candidate spatial anchors have a communicating laboratory path, and the two auxiliary record information and the personnel record information of the technicians do not conflict, then it is determined that there is a direct transfer relationship between the two adjacent sample events.

[0086] S1423. If the sum of the times of the two events does not cover the preset process time, the semantics of the two stages are not within the process flow template corresponding to the sample type, the two candidate spatial anchors do not have a communication laboratory path, or the two auxiliary record information and the personnel record information of the technicians conflict, then it is determined that there is no direct transfer relationship between the two adjacent sample events.

[0087] A direct transfer relationship can be determined from at least four aspects: time, process, route, and evidence, as detailed below: 1. In terms of time, determine whether the time interval between the previous event and the next event can cover the reasonable time required for the sample to complete the transfer, waiting or process stop between the corresponding spatial locations; 2. In terms of process, determine whether the stage semantics of the previous event and the stage semantics of the subsequent event conform to the detection process template corresponding to the sample type. 3. Regarding the path, determine whether there is a passable laboratory path between the candidate spatial anchor point corresponding to the previous event and the candidate spatial anchor point corresponding to the next event; 4. Regarding evidence, determine whether the direct transfer relationship conflicts significantly with AGV task records, equipment loading / unloading records, manual operation records, or abnormal records.

[0088] If there exists at least one set of candidate spatial anchor point combinations that can simultaneously satisfy the above judgment conditions, then it is determined that there is a link relationship between the adjacent event pairs that can be directly connected, and the time range between the two is treated as a continuous link.

[0089] If there is no candidate spatial anchor point combination that meets the conditions, or if the candidate spatial anchor point combination has obvious conflicts with the time window, process flow, path connectivity, or auxiliary record evidence, it is determined that a credible direct transfer relationship cannot be formed between the adjacent event pairs, and the position between the two is marked as a link breakpoint.

[0090] S15. If it is determined that there is a link breakpoint in the global optimal link, then the global optimal link is completed to obtain an enhanced link, and the enhanced link is time-restored to obtain a sample processing link.

[0091] If it is determined that there is a link breakpoint in the globally optimal link, the location of the link breakpoint and the missing content can be determined. Then, the globally optimal link can be completed based on the location of the link breakpoint and the missing content to obtain the enhanced link. Finally, the corresponding business time can be added to the enhanced link to complete the time restoration and obtain the sample processing link.

[0092] In one embodiment, the step of completing the globally optimal link to obtain an enhanced link, and performing time restoration on the enhanced link to obtain a sample processing link, may include the following sub-steps: S151. Determine the link breakpoint of the globally optimal link and construct a set of virtual segments using the link breakpoint, wherein the link breakpoint is an adjacent sample event pair whose consistency score does not meet the preset consistency threshold and does not meet the timing constraints, process constraints and path constraints.

[0093] S152. Select and complete segments from the set of virtual segments, wherein the complete segments are virtual segments that meet the preset reasonable score in terms of completion time reachability, process rationality and path connectivity, and completion rationality score, and each virtual segment is constructed based on the link breakpoint.

[0094] S153. Insert the completed segment into the globally optimal link to obtain an enhanced link, and insert the time slice corresponding to the sample processing data into the enhanced link to obtain a sample processing link.

[0095] After a link breakpoint is marked, a virtual segment completion process can be triggered. The completion process may include: reading the stage semantics, event time, candidate spatial anchors, sample type, and process flow template of events before and after the breakpoint; obtaining AGV task records, manual handover records, manual handling records, equipment loading / unloading records, and detection anomaly records within the time window corresponding to the breakpoint; generating breakpoint context information based on the stage relationship before and after the breakpoint, time window, path relationship, and auxiliary record evidence; and submitting the breakpoint as a breakpoint to be completed to the subsequent virtual segment generation process, which will then continue to complete the breakpoint type identification, candidate virtual segment generation, verification, scoring, and optimal virtual segment selection.

[0096] For example, in a scenario where concrete test block samples are transferred from the curing station to the testing equipment station, if the sample event time sequence chain records the sample registration event, the curing station entry event, and the testing completion event in sequence, but no curing event, transfer event, or equipment loading event is recorded between the curing station entry event and the testing completion event, then it is first determined whether the curing station entry event and the testing completion event can be directly connected. If, according to the process flow template, the curing stage and the testing completion stage usually involve curing exit, transfer, and equipment loading, and there are records of manual handling or equipment loading within the corresponding time window, then the link between the curing station entry event and the testing completion event is marked as a breakpoint, and a virtual segment completion process is triggered.

[0097] In one embodiment, the completion type corresponding to the link breakpoint may include missing transport completion, waiting for stationary completion, manual relay transport completion, and abnormal handling process completion. The completion method is determined in the subsequent virtual segment generation step based on the stage semantics, time window, path relationship, and auxiliary record information of the events before and after the breakpoint, and its cost is reflected in the global objective function through missing completion penalty term or abnormal inconsistency penalty term.

[0098] Specifically, when there is no direct feasible transition between adjacent observed events, it indicates that there may be missing events in the middle. The link is completed using rules generated from virtual segments. Let the set of virtual segments to be completed be as follows: ; The virtual segment can be a virtual transfer segment, a virtual dwell segment, a virtual manual intervention transfer segment, or a virtual anomaly handling segment. The process of completing the link based on virtual segment generation rules includes the following steps: The first step is to identify all adjacent event pairs marked as not directly transferable or to be completed for transfer based on the optimal sample link results, and to determine the adjacent event pairs as link breakpoints.

[0099] The second step is to determine the type of link breakpoint by combining the stage semantics, time difference, path relationship, AGV task records, manual operation records and exception records of the events before and after the breakpoint.

[0100] In one operation mode, the type of link breakpoint is determined by combining the stage semantics, time difference, path relationship, AGV task records, manual operation records, and exception records of events before and after the breakpoint. The specific steps for supplementing the type determination process are as follows: First, read the stage semantics, event time, selected spatial anchor point, sample type, and process flow template of the events before and after the breakpoint. Then, determine the breakpoint time window based on the event time before and after the breakpoint. Within the breakpoint time window, retrieve AGV task records, manual handover records, manual handling records, equipment loading / unloading records, equipment occupancy records, and exception records.

[0101] If there is a spatial change between events before and after the breakpoint, an allowed path exists between spatial anchor points before and after the breakpoint, and the breakpoint time window covers the corresponding transfer time, but there is a lack of corresponding actual transfer event records, then the breakpoint is determined to be a missing transfer situation. If the stages before and after the breakpoint are technologically permissible to connect, the breakpoint time window is significantly longer than the shortest transfer time, and there are buffer positions, inspection positions, maintenance positions, or equipment waiting positions near the spatial anchor points before and after the breakpoint, then the breakpoint is determined to be a waiting and dwelling situation. If the AGV task record is missing, interrupted, canceled, or inconsistent with the spatial relationship before and after the breakpoint, and there are records of manual handover, manual handling, or manual operation, then the breakpoint is determined to be a manual relay transfer situation. If there are records of detection failure, equipment abnormality, review, return, abnormal work order, or abnormal handling, then the breakpoint is determined to be an abnormal handling situation.

[0102] When the same link breakpoint simultaneously meets the discrimination conditions of two or more breakpoint types, multiple possible breakpoint types are retained, and candidate virtual segments are generated according to each possible breakpoint type to avoid the exclusion of correct completion results due to premature single classification.

[0103] The third step is to generate a set of candidate virtual segments Z for different breakpoint types.

[0104] If the breakpoint type is a missing transport scenario, then the selected spatial anchor points of the previous event and the subsequent event in the initial optimal sample link are used as candidate start and end points, and one or more candidate virtual transport segments are generated along the allowed passage path between them. If the breakpoint type is a waiting and dwelling scenario, then candidate virtual dwelling segments are generated in the buffer positions, inspection positions, maintenance positions, equipment waiting positions, or other legal dwelling positions near the spatial anchor points before and after the breakpoint. If the breakpoint type is a manual relay transport scenario, then the spatial anchor points before and after the breakpoint, the manual handover positions, or the start and end points indicated by the manual handling records are used as candidate endpoints, and candidate virtual manual intervention transport segments are generated along the manually passable path. If the breakpoint type is an abnormal handling scenario, then the detection equipment position, the equipment rear buffer position, or the abnormal sample temporary storage position are used as candidate start points, and the abnormal handling position, the review position, the return temporary storage position, or the re-inspection equipment position are used as candidate endpoints or candidate dwelling positions, generating candidate virtual abnormal handling segments. The candidate virtual segments generated under the above breakpoint types are merged to form a candidate virtual segment set Z.

[0105] The fourth step is to verify the time reachability, process rationality, and path connectivity of set Z.

[0106] Specifically, candidate virtual segments z can be obtained from the set Z, and time-accessibility regularization scores can be determined for each segment. Standardized scoring value for process rationality and path connectivity regularization score .

[0107] The standardized score is used in the verification phase to determine whether a candidate virtual segment z meets the minimum requirements for entering the subsequent rationality scoring. It can be reused in the scoring phase as a basis for supplementing the rationality ranking. For any candidate virtual segment z, its starting spatial anchor point is defined as... The final spatial anchor point is The start time is The end time is Then its duration can be expressed as follows: ; The time accessibility regularization score can be represented by the following formula: ; in, Indicates the expected duration of candidate virtual segment z. This represents the time deviation decay parameter. For virtual transfer segments, The shortest path length from origin to destination, permitted transfer methods, and minimum operation time can be used to determine the process; for virtual dwell segments and virtual manual intervention transfer segments, It can be determined by combining the necessary waiting time and manual operation time.

[0108] Standardized scoring value for process rationality Determined according to the preset process template and breakpoint type matching rules. After inserting the candidate virtual segment z between the event before the breakpoint and the event after the breakpoint, if the candidate virtual segment z completely matches the process template with the stages before and after the breakpoint, then... Take the higher score; if the candidate virtual segment z belongs to the allowed waiting, caching, transit, manual intervention, or anomaly handling connection for this breakpoint type, then Take the median score; if the candidate virtual segment z has only a weak process association but is supported by auxiliary records, then... Take the lower score; if the candidate virtual segment z violates the process flow sequence and there are no abnormal records, manual review records, or other supporting records, then... The result is set to 0, and it is determined that the process rationality verification has not been passed.

[0109] The path connectivity regularization score can be represented by the following formula: ; The starting and ending spatial anchor points of the candidate virtual segment z are represented in the laboratory path diagram. The value is 1 if an allowed path exists, and 0 if it does not. It consists of a set of spatial anchor nodes and a set of traversable path edges. Spatial anchor nodes include workstations, rack positions, buffer positions, equipment positions, handover positions, and exception handling positions; traversable path edges include AGV track segments, manual passage segments, equipment entry and exit buffer path segments, and inter-area connecting paths. A traversable path edge must at least include path length, permitted passage mode, permitted sample type, passage direction, area it belongs to, and passage status. Indicates from to The path travel cost, according to the laboratory path map This is permitted in China.

[0110] The path length, estimated travel time, necessary operation time, matching degree of travel mode, and path status are comprehensively determined for the travel route. If there are multiple allowed travel paths between the starting spatial anchor point and the ending spatial anchor point, the path with the lowest travel cost is selected as the path basis for the candidate virtual segment. This represents the path cost attenuation parameter.

[0111] When the candidate virtual segment z simultaneously satisfies the following conditions , , If all values ​​are greater than or equal to the minimum verification threshold, the candidate virtual segment z is added to the set of candidate virtual segments that have passed the verification.

[0112] Step 5: Calculate the completion rationality score by combining the event context before and after the breakpoint; for candidate virtual segments z that pass the verification, further calculate the completion rationality score. The completion rationality score is expressed as: ; in , and These are the aforementioned standardized scoring values, which are reused as the ranking criteria during the scoring phase. a1 to a5 represent the corresponding weights. This represents the support score for auxiliary records, used to measure the degree to which external business records such as AGV task records, manual operation records, abnormal work orders, equipment alarm records, and manual review records support the candidate virtual segment z. The support score for auxiliary records is determined according to a preset rule table; a high score is given when a direct auxiliary record exists that is identical to the candidate virtual segment z in terms of start point, end point, time window, and segment type; a medium score is given when some fields are identical or auxiliary records indirectly support the candidate virtual segment; a low score is given when only weakly related auxiliary records exist; and a score of 0 is given when no supportable auxiliary records exist. (Relative conflict penalty item) A zero value is used when there is no identifiable conflict between the candidate virtual segment and the events before and after the breakpoint; a low penalty value is used when there is a slight time deviation, slight path detour, or weak process inconsistency but it does not reach the level of elimination; and a high penalty value is used when there is an obvious conflict that can be explained by anomaly records, manual review records, or equipment logs.

[0113] Step 6: Among the candidate virtual segments that have passed the verification, select the virtual segment with the highest completion rationality score and the best overall consistency with the preceding and following events as the completion result of the current link breakpoint.

[0114] Specifically, to determine whether a segment is the virtual segment with the best overall consistency with the preceding and following events, the candidate virtual segments that pass the verification can be sorted according to the completion rationality score, and the candidate virtual segments with the highest score or the closest score are selected to form the preferred candidate set.

[0115] Each candidate virtual segment in the preferred candidate set is inserted between the event before the breakpoint and the event after the breakpoint, and it is determined whether the insertion will cause new inconsistencies in the original link.

[0116] Inconsistencies include: the starting point of the candidate virtual segment cannot connect to the spatial anchor point of the event before the breakpoint; the ending point of the candidate virtual segment cannot connect to the spatial anchor point of the event after the breakpoint; the insertion results in new time holes, time overlaps, or time inversions; the insertion causes inconsistencies between the sample stage sequence and the process flow template; the insertion causes path jumps, cross-regional jumps, or path discontinuities; the insertion causes the same AGV task record, manual operation record, equipment record, or abnormal record to be interpreted repeatedly, or to conflict with evidence of confirmed real events.

[0117] In the preferred candidate set, if a candidate virtual segment does not generate any new inconsistencies after insertion, or generates the fewest new inconsistencies with the lowest severity and the fewest required explanation assumptions, then the candidate virtual segment is determined to have the best overall consistency with the events before and after the breakpoint, and is selected as the completion result for the current link breakpoint. If multiple candidate virtual segments have the same overall consistency, the candidate virtual segment with the highest completion rationality score is selected.

[0118] Step 7: Insert the candidate virtual segments into the original optimal sample link in the corresponding order to form an enhanced link; combine the enhanced link with the original optimal sample link to form an enhanced link. It can be represented as: ; Step 8: Record the source of each virtual segment's breakpoint, the reason for its generation, the preceding and following events, the completion type, and the confidence level for subsequent time-slice state restoration and 3D scene linkage display. Through the above steps, virtual segment completion is gradually completed based on link breakpoint identification, breakpoint type judgment, candidate virtual segment generation, constraint verification, optimal selection, enhanced link insertion, and completion description recording.

[0119] Finally, for any time slice In enhancing the link The segment covered by the mid-position at that moment .like If it is a stationary or detection segment, the current position is determined by the corresponding fixed spatial anchor point; if it is a transfer or abnormal segment, the current position is calculated based on the segment's start and end times and path length. ; in: and These represent the spatial locations of the starting and ending points of the segment, respectively. and These represent the start and end times of the segment, respectively. Also output: ; in Indicates the current position. Indicates the current stage of the process. This indicates the observed event or virtual segment that dominates the judgment at that moment. Indicates the cause of the abnormality or completes the reason.

[0120] In one embodiment, existing technologies employ 3D visualization technology to visualize the data link after its construction. For example, a digital twin system for a testing laboratory typically uses 3D visualization technology to construct a laboratory space scene and graphically display objects such as equipment, workstations, shelves, and AGV routes. 3D visualization technology can achieve effects such as displaying the overall laboratory layout, locating equipment objects, highlighting prompts, playing animations, and displaying equipment status. However, 3D visualization capabilities focus primarily on scene representation, object presentation, and interactive displays, and do not directly address the problem of reconstructing the actual data link experienced by the sample.

[0121] To improve visualization quality, as an example, after the steps of completing the globally optimal link to obtain the enhanced link and performing time restoration on the enhanced link to obtain the sample processing link, the method may further include the following steps: S16. Extract the segment information corresponding to the sample event in the time segment from the sample processing link, wherein the segment information includes: spatial anchor point, path segment, segment type, start and end time, current position, stage to which it belongs, and abnormal description information.

[0122] S17. Visualize the segment information, wherein the visualization process includes visually displaying the segment information in chronological order, locating real or virtual segments in the visualization sequence, and adjusting the animation and style according to the segment type.

[0123] Specifically, spatial anchors and path segments in the enhanced link can be bound to a collection of 3D scene objects.

[0124] The collection of 3D scene objects must include at least the following object types: 1. Spatial node objects corresponding to the sample receiving station, sample sorting station, temporary storage rack station, curing station, buffer station, testing equipment station, manual handover station, and abnormal handling station.

[0125] 2. Path objects corresponding to AGV track sections, manual passage sections, and equipment entry / exit buffer path sections.

[0126] 3. Label objects, status panel objects, and descriptions corresponding to the current location, current stage, cause of the anomaly, and time information of the sample.

[0127] 4. Style objects used to represent the display status of historical segments, currently active segments, future segments to be executed, and abnormal segments.

[0128] In one embodiment, the process of binding spatial anchor points and path segments to a collection of 3D scene objects includes the following steps: 1.1. Establish unique object codes for each node object and path object in the 3D scene in advance, establish the mapping relationship between spatial anchor point identifiers and 3D node object codes, and the mapping relationship between path segment identifiers and 3D path object codes.

[0129] 2.2. Establish an object attribute table for each 3D scene object. The object attribute table shall include at least the object code, object type, corresponding business space identifier, 3D coordinates or path point set, region, displayable status, bindable attributes and interactive response method.

[0130] 3. Read the enhanced link results and extract the anchor sequence, path segment sequence, segment type and corresponding time range.

[0131] 4. Based on the above mapping relationship, convert the spatial anchor points and path segments in the enhanced link into corresponding 3D node objects and path objects.

[0132] 5. Add display attributes such as segment type, display status, color style, time range, and descriptive information to the converted 3D object.

[0133] 6. Output object binding results that can be directly called by the 3D scene for subsequent highlighting, shading, translation animation, and interactive display of the description panel. The object binding results should include at least one or more of the following: object code, object type, corresponding segment type, start and end time, display style, description information, and exception identifier.

[0134] In one operational mode, the 3D display process involves at least six stages: enhanced link result output, display sequence encapsulation, object matching, style calculation, time-driven update, and explanatory linkage output. This process performs temporal, object-oriented, and visual transformations on the link recovery results. The process of displaying a set of 3D scene objects to form a sample spatiotemporal link includes the following steps: 1. Read the enhanced link and arbitrary time slice state restoration results to obtain the spatial anchor point, path segment, segment type, start and end time, current position, stage, and anomaly description information of the sample in each time segment.

[0135] 2. Organize the above results into a display sequence according to the time order. The display sequence shall include at least the object code, object type, segment type, start and end time, display status, color style, explanatory text and exception indicator.

[0136] 3. Based on the current playback time or query time, locate the corresponding real or virtual segment in the display sequence.

[0137] 4. If the current segment is a stationary segment or a detection segment, the corresponding 3D node object will be driven to perform highlighting, coloring, blinking, or label display; if the current segment is a transfer segment, a manually intervened transfer segment, or an anomaly handling segment, the corresponding 3D path object will be driven to perform path coloring, flow effects, or displacement animation.

[0138] 5. Assign different display styles to historically completed segments, currently active segments, future segments to be executed, and abnormal segments to reflect the temporal sequence of the link and abnormal status.

[0139] 6. Synchronously update the status panel, timeline, current position label, stage description, and anomaly explanation to ensure that the object display results in the 3D scene are consistent with the enhanced link interpretation results.

[0140] 7. When dragging, playing, pausing, fast forwarding, rewinding, or querying at a specified time on the timeline, the current segment positioning, object-driven, and explanatory information linkage steps are repeatedly executed to form a complete sample spatiotemporal link display process.

[0141] Reference Figure 2 The diagram illustrates an operation flowchart of a link construction method for a dark laboratory sample according to an embodiment of the present invention.

[0142] Specifically, the operation process of the link construction method for the dark laboratory sample may include the following steps: The first step is to collect raw data from multiple sources to obtain sample processing data for samples from the dark laboratory.

[0143] The second step is standard sample event normalization, which involves standardizing the raw data from multiple sources.

[0144] The third step is to construct a sample event time series chain, which is built based on the standardized multi-source raw data.

[0145] The fourth step is candidate spatial anchor generation and credibility scoring. Candidate spatial anchors are generated based on the sample event time series chain, and the credibility score of each candidate spatial anchor is calculated.

[0146] The fifth step is the joint link recovery of three constraints. Based on the timing constraints, process constraints, path constraints and credibility scores, the global optimal link is generated by using candidate spatial anchor points.

[0147] The sixth step is to complete the virtual segment and form the enhanced link. The enhanced link is obtained by completing the globally optimal link based on the virtual segment.

[0148] Step 7: Restore the time slice state and add the time slice to the enhanced link.

[0149] Step 8: 3D object binding and display, providing a visual representation of the enhanced link.

[0150] This invention enables the unified expression of heterogeneous data in laboratory management systems, equipment systems, AGV systems, and manual records by constructing standard sample event objects and performing stage semantic regularization, thus providing a consistent data foundation for subsequent link recovery.

[0151] This invention can improve the robustness and interpretability of sample spatial localization by first generating a set of candidate spatial anchor points and then performing a confidence score screening, thus avoiding the direct and rigid mapping of incomplete events to a single location.

[0152] This invention can comprehensively determine the possible transfer relationships between adjacent events by jointly introducing timing constraints, process constraints and path constraints, thereby enabling the selection of sample experience links that better conform to the actual operation logic of the laboratory from multiple candidate paths.

[0153] This invention can introduce a virtual segment completion mechanism and an anomaly recovery mechanism to enable samples to form a continuous and well-explained enhanced link even in cases of missing intermediate records, failed detections, or manual relay transport.

[0154] This invention can output the sample location, stage, supporting events, and cause of anomalies at a specified time through an arbitrary time slice state restoration mechanism, providing a unified basis for sample traceability, anomaly verification, and three-dimensional linkage display.

[0155] In this embodiment, the present invention provides a method for constructing a link for samples from a dark laboratory. The beneficial effects are as follows: The present invention can acquire sample processing data of samples from a dark laboratory and construct a time series chain of sample events using this data; calculate the confidence score for each sample event in the time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and the consistency score value; filter the events in the time series chain based on the confidence score value and the consistency score value and construct a globally optimal link for the sample events; determine whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events; if a link breakpoint is determined to exist in the globally optimal link, complete the globally optimal link to obtain an enhanced link, and perform time restoration on the enhanced link to obtain the sample processing link. When constructing the link, the present invention can determine whether there is a link breakpoint in the globally optimal link based on the transition relationship between two sample events, and complete the link when a breakpoint exists, which can avoid missing link data and ensure the accuracy and precision of the link.

[0156] This invention also provides a link construction device for samples in a dark laboratory, see [link to relevant documentation]. Figure 3 The diagram shows a schematic of a link construction device for a dark laboratory sample according to an embodiment of the present invention.

[0157] As an example, the link construction device for the dark laboratory sample may include: The acquisition module 201 is used to acquire sample processing data of samples in a dark laboratory and to construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and record business processing data. The calculation module 202 is used to calculate the confidence score corresponding to each sample event in the sample event time series chain and the consistency score corresponding to two adjacent sample events respectively, so as to obtain the confidence score value and the consistency score value. The filtering module 203 is used to filter the events in the sample event time series chain according to the credibility score and the consistency score and to construct the globally optimal link of the sample events. The determination module 204 is used to determine whether there is a link breakpoint in the global optimal link based on the transfer relationship between two adjacent sample events in the global optimal link; The construction module 205 is used to complete the globally optimal link to obtain an enhanced link if it is determined that there is a link breakpoint in the globally optimal link, and to perform time restoration on the enhanced link to obtain a sample processing link.

[0158] Optionally, the step of calculating the confidence score for each sample event in the sample event time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and the consistency score value includes: Based on a preset set of fixed spatial anchor points and a preset set of transport path segments, each sample event in the sample event time series chain is converted into a candidate spatial anchor point and combined into a candidate spatial anchor point set. The candidate spatial anchor point set includes multiple candidate spatial anchor points, and each candidate spatial anchor point corresponds to a sample event. Determine the credibility parameter corresponding to each candidate spatial anchor point, and calculate the credibility score using the credibility parameters corresponding to two adjacent candidate spatial anchor points. The credibility parameter includes: comprehensive rule-based score, event source reliability factor, and global conflict penalty item. Determine the consistency parameters corresponding to each candidate spatial anchor point, and calculate the consistency score using the consistency parameters corresponding to two adjacent candidate spatial anchor points. The consistency parameters include: timing constraints, process constraints, path constraints, and conflict penalty terms.

[0159] Optionally, the step of filtering the events in the sample event time series chain based on the confidence score and the consistency score, and constructing the globally optimal link of the sample events, includes: Based on the credibility score, candidate spatial anchors that meet the preset credibility threshold are selected from the candidate spatial anchor set corresponding to the sample event time series chain as filter anchors. Based on the consistency score, consistent anchors and inconsistent anchors are selected from the filter anchors. The consistent anchors are filter anchors whose consistency scores meet a preset consistency threshold, and the inconsistent anchors are filter anchors whose consistency scores do not meet the preset consistency threshold but meet timing constraints, process constraints, and path constraints. A hierarchical extended state diagram is constructed using the consistent anchor points and the inconsistent anchor points, and the hierarchical extended state diagram is solved to obtain the globally optimal link.

[0160] Optionally, determining whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events in the globally optimal link includes: The processing information for identifying two adjacent sample events within the globally optimal link includes: event time, stage semantics, candidate spatial anchor point, sample type, process flow template, and auxiliary record information. Based on the processing information, determine whether there is a direct transfer relationship between two adjacent sample events; If it is determined that there is a direct transfer relationship between every two adjacent sample events, then it is determined that there is no link breakpoint in the global optimal link; If it is determined that there is no direct transfer relationship between any two adjacent sample events, then it is determined that there is a link breakpoint in the globally optimal link.

[0161] Optionally, determining whether there is a direct transfer relationship between two adjacent sample events based on the processing information includes: Each of the following criteria is determined: whether the sum of the times of the two events covers the preset process time; whether the semantics of the two stages are within the process flow template corresponding to the sample type; whether there is a communication laboratory path between the two candidate spatial anchors; and whether there is an information conflict between the two auxiliary record information and the personnel record information of the technicians. If the sum of the times of the two events covers the preset process time, the semantics of the two stages are within the process flow template corresponding to the sample type, the two candidate spatial anchors have a communicating laboratory path, and the two auxiliary record information and the personnel record information of the technicians do not conflict, then it is determined that there is a direct transfer relationship between the two adjacent sample events. If the sum of the times of the two events does not cover the preset process time, the semantics of the two stages are not within the process flow template corresponding to the sample type, the two candidate spatial anchors do not have a communication laboratory path, or the two auxiliary record information conflicts with the personnel record information of the technicians, then it is determined that there is no direct transfer relationship between the two adjacent sample events.

[0162] Optionally, the step of completing the globally optimal link to obtain an enhanced link, and performing time restoration on the enhanced link to obtain a sample processing link, includes: Determine the link breakpoint of the globally optimal link and construct a set of virtual segments using the link breakpoint, wherein the link breakpoint is an adjacent sample event pair whose consistency score does not meet the preset consistency threshold and does not meet the timing constraints, process constraints and path constraints; The virtual segment set is used to filter and complete segments, wherein the completed segments are virtual segments that meet the preset reasonable score in terms of completion time reachability, process rationality and path connectivity, and the completion rationality score is satisfied. Each virtual segment is constructed based on the link breakpoint. The completed segment is inserted into the globally optimal link to obtain an enhanced link, and the time slice corresponding to the sample processing data is inserted into the enhanced link to obtain a sample processing link.

[0163] Optionally, the device further includes: The extraction module is used to extract the segment information corresponding to the sample event in the time segment from the sample processing link after the steps of completing the global optimal link to obtain the enhanced link and performing time restoration on the enhanced link to obtain the sample processing link. The segment information includes: spatial anchor point, path segment, segment type, start and end time, current position, stage, and anomaly description information. The visualization module is used to visualize the segment information, wherein the visualization process includes visually displaying the segment information in chronological order, locating real or virtual segments in the visualization sequence, and adjusting the animation and style according to the segment type.

[0164] Those skilled in the art will understand that, for ease of description and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0165] Furthermore, this application also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the link construction method for a black-light laboratory sample as described in the above embodiments.

[0166] Furthermore, embodiments of this application also provide a computer-readable storage medium storing a computer-executable program for causing a computer to execute the link construction method for a black-light laboratory sample as described in the above embodiments.

[0167] In the description of the embodiments of the present invention, it should be noted that the terms "above," "below," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. When an element such as a layer, region, or substrate is referred to as being "above" or "on top of" another element, it may be directly on the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly on" or "above" another element, there is no intermediate element. It should also be understood that when an element is referred to as being "below" or "under" another element, it may be directly below or under the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly below" or "under" another element, there is no intermediate element. Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0168] Those skilled in the art will understand that embodiments of this application may also include computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0169] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), devices, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0170] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0171] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0172] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for constructing a link for a sample in a dark laboratory, characterized in that, The method includes: Acquire sample processing data of samples from a dark laboratory, and construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and recorded business processing data; The credibility score and the consistency score of each sample event in the sample event time series chain are calculated respectively to obtain the credibility score value and the consistency score value. The events in the sample event time series chain are filtered based on the credibility score and the consistency score, and a globally optimal link for the sample events is constructed. Determine whether there is a link breakpoint in the global optimal link based on the transition relationship between two adjacent sample events in the global optimal link; If it is determined that there is a link breakpoint in the globally optimal link, then the globally optimal link is completed to obtain an enhanced link, and the enhanced link is time-restored to obtain a sample processing link.

2. The method for constructing a link for a sample in a dark laboratory according to claim 1, characterized in that, The process of calculating the confidence score for each sample event in the sample event time series chain and the consistency score for two adjacent sample events to obtain the confidence score value and consistency score value includes: Based on a preset set of fixed spatial anchor points and a preset set of transport path segments, each sample event in the sample event time series chain is converted into a candidate spatial anchor point and combined into a candidate spatial anchor point set. The candidate spatial anchor point set includes multiple candidate spatial anchor points, and each candidate spatial anchor point corresponds to a sample event. Determine the credibility parameter corresponding to each candidate spatial anchor point, and calculate the credibility score using the credibility parameters corresponding to two adjacent candidate spatial anchor points. The credibility parameter includes: comprehensive rule-based score, event source reliability factor, and global conflict penalty item. Determine the consistency parameters corresponding to each candidate spatial anchor point, and calculate the consistency score using the consistency parameters corresponding to two adjacent candidate spatial anchor points. The consistency parameters include: timing constraints, process constraints, path constraints, and conflict penalty terms.

3. The method for constructing a link for a sample in a dark laboratory according to claim 2, characterized in that, The step of filtering the events in the sample event time series chain based on the credibility score and the consistency score, and constructing the globally optimal link of the sample events, includes: Based on the credibility score, candidate spatial anchors that meet the preset credibility threshold are selected from the candidate spatial anchor set corresponding to the sample event time series chain as filter anchors. Based on the consistency score, consistent anchors and inconsistent anchors are selected from the filter anchors. The consistent anchors are filter anchors whose consistency scores meet a preset consistency threshold, and the inconsistent anchors are filter anchors whose consistency scores do not meet the preset consistency threshold but meet timing constraints, process constraints, and path constraints. A hierarchical extended state diagram is constructed using the consistent anchor points and the inconsistent anchor points, and the hierarchical extended state diagram is solved to obtain the globally optimal link.

4. The method for constructing a link for a sample in a dark laboratory according to claim 1, characterized in that, The step of determining whether there is a link breakpoint in the globally optimal link based on the transition relationship between two adjacent sample events in the globally optimal link includes: The processing information for identifying two adjacent sample events within the globally optimal link includes: event time, stage semantics, candidate spatial anchor point, sample type, process flow template, and auxiliary record information. Based on the processing information, determine whether there is a direct transfer relationship between two adjacent sample events; If it is determined that there is a direct transfer relationship between every two adjacent sample events, then it is determined that there is no link breakpoint in the global optimal link; If it is determined that there is no direct transfer relationship between any two adjacent sample events, then it is determined that there is a link breakpoint in the globally optimal link.

5. The method for constructing a link for a sample in a dark laboratory according to claim 4, characterized in that, The step of determining whether there is a direct transfer relationship between two adjacent sample events based on the processing information includes: Each of the following criteria is determined: whether the sum of the times of the two events covers the preset process time; whether the semantics of the two stages are within the process flow template corresponding to the sample type; whether there is a communication laboratory path between the two candidate spatial anchors; and whether there is an information conflict between the two auxiliary record information and the personnel record information of the technicians. If the sum of the times of the two events covers the preset process time, the semantics of the two stages are within the process flow template corresponding to the sample type, the two candidate spatial anchors have a communicating laboratory path, and the two auxiliary record information and the personnel record information of the technicians do not conflict, then it is determined that there is a direct transfer relationship between the two adjacent sample events. If the sum of the times of the two events does not cover the preset process time, the semantics of the two stages are not within the process flow template corresponding to the sample type, the two candidate spatial anchors do not have a communication laboratory path, or the two auxiliary record information conflicts with the personnel record information of the technicians, then it is determined that there is no direct transfer relationship between the two adjacent sample events.

6. The method for constructing a link for a sample in a dark laboratory according to any one of claims 1-5, characterized in that, The process of completing the globally optimal link to obtain an enhanced link, and then performing time restoration on the enhanced link to obtain a sample processing link, includes: Determine the link breakpoint of the globally optimal link and construct a set of virtual segments using the link breakpoint, wherein the link breakpoint is an adjacent sample event pair whose consistency score does not meet the preset consistency threshold and does not meet the timing constraints, process constraints and path constraints; The virtual segment set is used to filter and complete segments, wherein the completed segments are virtual segments that meet the preset reasonable score in terms of completion time reachability, process rationality and path connectivity, and the completion rationality score is satisfied. Each virtual segment is constructed based on the link breakpoint. The completed segment is inserted into the globally optimal link to obtain an enhanced link, and the time slice corresponding to the sample processing data is inserted into the enhanced link to obtain a sample processing link.

7. The method for constructing a link for a sample in a dark laboratory according to claim 6, characterized in that, After the steps of completing the globally optimal link to obtain the enhanced link and performing time restoration on the enhanced link to obtain the sample processing link, the method further includes: Extract the segment information corresponding to the sample event in the time segment from the sample processing link, wherein the segment information includes: spatial anchor point, path segment, segment type, start and end time, current position, stage to which it belongs and anomaly description information; The segment information is visualized, which includes visualizing the segment information in chronological order, locating real or virtual segments in the visualization sequence, and adjusting the animation and style according to the segment type.

8. A link construction device for samples in a dark laboratory, characterized in that, The device includes: The acquisition module is used to acquire sample processing data of samples from a dark laboratory and to construct a sample event time series chain using the sample processing data, wherein the sample processing data includes sample feature data and record business processing data. The calculation module is used to calculate the confidence score corresponding to each sample event in the sample event time series chain and the consistency score corresponding to two adjacent sample events, so as to obtain the confidence score value and the consistency score value. The filtering module is used to filter the events in the sample event time series chain based on the credibility score and the consistency score, and to construct the globally optimal link of the sample events. The determination module is used to determine whether there is a link breakpoint in the global optimal link based on the transfer relationship between two adjacent sample events in the global optimal link; The module is configured to, if it is determined that there is a link breakpoint in the globally optimal link, complete the globally optimal link to obtain an enhanced link, and perform time restoration on the enhanced link to obtain a sample processing link.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the link construction method for a dark laboratory sample as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the link construction method for a black-light laboratory sample as described in any one of claims 1-7.

Images