A timing data alignment method and system for a stainless steel pot production line

Abstract

This application provides a method and system for aligning time-series data of a stainless steel pot production line. The method includes: acquiring production data from multiple workstations of the stainless steel pot production line within a target time period and converting the production data into multiple event nodes; extracting the identity features of the stainless steel pots from the multiple event nodes; establishing candidate transfer relationships among the multiple event nodes based on the process topology relationship, flow time constraints, and identity features between the multiple workstations to form a candidate transfer network; solving the trajectory of the candidate transfer network to obtain the production history trajectory corresponding to each of the multiple stainless steel pots within the target time period; and correcting the production data with abnormal time-series relationships within the target time period based on the production history trajectory.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for aligning time-series data in a stainless steel pot production line. Background Technology

[0002] In stainless steel pot production lines, due to the lack of continuously readable explicit identification for individual pots, the mixing of multiple pots of the same specifications in the buffer area, rework and order insertion disturbances, coupled with the diverse data sources of each workstation, asynchronous time, and delayed uploads, it is difficult to accurately attribute the production data collected by each workstation to specific pots. This makes it impossible to effectively reconstruct the complete production process of a single pot, affecting the accuracy of product quality traceability and process optimization. Summary of the Invention

[0003] In view of this, this application provides a timing data alignment method, system and electronic equipment for a stainless steel pot production line to overcome the shortcomings of the prior art.

[0004] According to a first aspect of this application, a method for aligning time-series data of a stainless steel pot production line is provided, comprising: acquiring production data of multiple workstations of the stainless steel pot production line within a target time period, and converting the production data into multiple event nodes; extracting the identity features of stainless steel pot bodies from the multiple event nodes; establishing candidate transfer relationships among the multiple event nodes based on the process topology relationship, flow time constraints, and identity features between the multiple workstations, forming a candidate transfer network; performing trajectory solving on the candidate transfer network to obtain the production history trajectory corresponding to each of the multiple stainless steel pot bodies within the target time period; and correcting production data with abnormal time-series relationships within the target time period based on the production history trajectory.

[0005] The second aspect of this application provides a time-series data alignment system for a stainless steel pot production line, comprising: an event node generation module for acquiring production data from multiple workstations within a target time period and converting the production data into multiple event nodes; an identity feature extraction module for extracting identity features of the stainless steel pots from the multiple event nodes; a candidate network construction module for establishing candidate transfer relationships among the multiple event nodes based on the process topology relationship, flow time constraints, and identity features between the multiple workstations, forming a candidate transfer network; a production trajectory solving module for solving the trajectory of the candidate transfer network to obtain the production history trajectory corresponding to each of the multiple stainless steel pots within the target time period; and a time-series data alignment module for correcting production data with abnormal time-series relationships within the target time period based on the production history trajectory.

[0006] A third aspect of this application 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 computer program to implement a timing data alignment method for any of the stainless steel pot production lines described above.

[0007] By adopting the technical solution of this application, the dispersed production data of multiple workstations on a stainless steel pot production line within a target time period is converted into multiple event nodes, and the identity features of the stainless steel pots are extracted from them. In the case that a single pot lacks a continuously readable explicit identifier, implicit features are used to construct a distinguishable basis for each pot. Then, based on the process topology relationship, flow time constraints, and identity features between multiple workstations, candidate transfer relationships are established between event nodes to form a candidate transfer network. Under complex conditions such as buffer mixing, rework, order insertion disturbances, and data time asynchrony and delayed upload, multiple possible data attribution relationships are expressed in a structured way. By solving the trajectory of the candidate transfer network, the production history trajectory corresponding to each stainless steel pot is identified from multiple candidate paths, realizing the accurate attribution of dispersed production data to specific pot individuals, effectively restoring the complete production process of a single pot, and further improving the accuracy and consistency of the data by correcting production data with abnormal time sequence relationships.

[0008] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0009] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which: Figure 1 This is a flowchart illustrating a timing data alignment method for a stainless steel pot production line provided in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a timing data alignment system for a stainless steel pot production line provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0010] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0011] Figure 1 A flowchart illustrating a timing data alignment method for a stainless steel pot production line provided in an embodiment of this application is shown.

[0012] like Figure 1 As shown, the timing data alignment method for the stainless steel pot production line includes steps S101 to S105.

[0013] Step S101: Obtain production data from multiple workstations on the stainless steel pot production line within the target time period, and convert the production data into multiple event nodes; Step S102: Extract the identity features of the stainless steel pot body from multiple event nodes; Step S103: Based on the process topology relationship, flow time constraints and identity characteristics between multiple workstations, establish candidate transfer relationships between multiple event nodes to form a candidate transfer network; Step S104: Solve the trajectory of the candidate transfer network to obtain the production history trajectory of each of the multiple stainless steel pots within the target time period. Step S105: Based on the production history trajectory, correct the production data that has abnormal time sequence relationships within the target time period.

[0014] In step S101, the stainless steel pot production line refers to an automated or semi-automated production line used to produce stainless steel pot products, including all processing stages from raw material input to finished product output. The stainless steel pot production line may include multiple workstations; a workstation refers to a work position on the production line where a specific process is performed. Each workstation is equipped with corresponding processing equipment or testing devices, and the pot body flows sequentially between workstations to complete the processing.

[0015] Optionally, the workstations of the stainless steel pot production line may include, but are not limited to: stamping workstation, welding workstation, polishing workstation, sandblasting workstation, inspection workstation, and packaging workstation.

[0016] Production data at a workstation refers to the information records related to pot body processing generated or collected at the workstation during the production process. These records are used to document key information such as the processing time, processing parameters, and quality status of the pot body at that workstation.

[0017] Optionally, the production data of the workstation may include, but is not limited to: workstation identification, timestamp, pot body specifications and model, processing parameters, quality inspection results, operator identification, equipment status information, etc.

[0018] Event nodes refer to data units formed by structuring production data at workstations, which are used to establish relationships between data from different workstations in subsequent analysis.

[0019] In one feasible implementation, the production data of each workstation can be continuously collected by the production line's data acquisition system within the target time period. The collected raw data is parsed and extracted according to dimensions such as workstation, time, and pot body characteristics. A corresponding event node is generated for each piece of production data. The event node contains structured fields such as workstation identifier, occurrence time, and pot body characteristic attributes.

[0020] It should be noted that, due to the diverse sources of data from each workstation and the existence of upload delays, the original timestamps of the data are retained when converting them into event nodes, and the time information is not modified, so that correlation analysis can be performed based on time constraints in subsequent steps.

[0021] In step S102, the identity feature refers to the attribute information that can be used to distinguish or identify different stainless steel pot bodies. It can be understood as the feature identifier carried by the pot body during the production process that can be used for identity determination, and is used to determine whether different event nodes correspond to the same pot body in the case of multiple pot bodies being mixed.

[0022] Optionally, the identification features may include, but are not limited to: pot body specifications and model, material type, batch identification, order number, pot body size parameters, surface feature code, RFID tag information, QR code information, etc.

[0023] In one feasible implementation, explicit identity features such as pot specifications, batch identification, and order number can be directly read from the structured fields of the event node. For event nodes containing RFID tag information or QR code information, the corresponding unique identifier is extracted as an identity feature. For event nodes containing only pot size parameters or material type, these parameters are combined as implicit identity features.

[0024] In step S103, the process topology relationship refers to the sequence of processes and the allowed flow paths between each workstation on the stainless steel pot production line.

[0025] Optionally, the process topology can be defined according to the standard process flow of the stainless steel pot production line, including normal sequential flow paths and rework return paths. For example, the flow from the stamping station to the welding station is normal flow, and the flow from the inspection station back to the polishing station is rework flow.

[0026] The flow time constraint refers to the time range required for the pot to flow between adjacent workstations, and is used to determine whether two event nodes may belong to the flow process of the same pot in terms of time.

[0027] Optionally, the flow time constraint can be obtained from historical production data statistics, including minimum flow time and maximum flow time. The minimum flow time reflects the physical transfer time of the pot between workstations, and the maximum flow time reflects the upper limit of a reasonable duration including buffer waiting.

[0028] Candidate transfer relationship refers to the possible correspondence between two event nodes in the flow of the pot, which is used to express the potential association between data records generated by a pot at different workstations.

[0029] A candidate transition network refers to a directed graph structure consisting of multiple event nodes and their candidate transition relationships. It can be understood as a set representation of all possible pot body transfer paths within a target time period.

[0030] In one feasible implementation, multiple event nodes can be traversed. For each event node as the starting node, subsequent workstation event nodes that satisfy the process topology relationship are selected from the remaining event nodes. For the selected event nodes, it is further determined whether the difference between its timestamp and the timestamp of the starting node is within the flow time constraint range. For event nodes that satisfy the time constraint, its identity characteristics are compared with those of the starting node to see if they are consistent or compatible. When the identity characteristics are consistent or compatible, a candidate transfer relationship is established between the starting node and the event node. All established candidate transfer relationships are summarized to form a candidate transfer network.

[0031] In another feasible implementation, event nodes can be grouped by workstation according to the process topology. For each workstation group, the event nodes in the group are sorted by timestamp. For each event node in a certain workstation group, the event nodes with timestamps later than the event node are searched in the subsequent workstation groups allowed by the process topology. The time difference is calculated and compared with the flow time constraint. For event node pairs that meet the time constraint, identity feature matching judgment is performed. When the key attributes such as specifications and batch identifiers in the identity features are the same, a candidate transfer relationship is established. The candidate transfer relationships established between each workstation group are connected to form a candidate transfer network.

[0032] It should be noted that due to buffer mixing, rework, and order insertion disturbances, an event node may establish candidate transfer relationships with multiple subsequent event nodes, forming multiple candidate paths. The candidate transfer network contains these multiple possible paths.

[0033] In step S104, the production history trajectory refers to the sequence of workstations and corresponding event nodes that a single pot body passes through sequentially within the target time period.

[0034] Alternatively, the trajectory can be solved using a path search algorithm in graph theory to find the optimal or most reasonable combination of paths while satisfying process and time constraints.

[0035] Optionally, the production history track can include detailed information such as the dwell time of the pot body at each workstation, processing parameters, quality inspection results, and time spent transferring between workstations, forming a complete production process record.

[0036] In one feasible implementation, the candidate transition network can be modeled as a directed graph, with event nodes as vertices and candidate transition relationships as edges. A path search algorithm is used to find a complete path from the starting station to the ending station. If an event node has multiple outgoing edges, the optimal outgoing edge is selected based on the scores of subsequent event nodes and the current event node in terms of identity feature matching degree and time interval rationality. During the path search process, it is ensured that each event node is used by only one trajectory to avoid multiple trajectories sharing the same event node. The multiple complete paths obtained by the search are used as production history trajectories for different pot bodies.

[0037] It should be noted that when there is a conflict in the candidate transfer network, that is, when multiple starting event nodes point to the same target event node, it is necessary to judge based on indicators such as identity feature matching degree and time proximity, and assign the target event node to the most matching starting event node. The unselected candidate transfer relationship will be eliminated.

[0038] In step S105, abnormal timing relationship refers to the time sequence relationship in the production data that does not conform to the actual process flow or physical laws.

[0039] Optionally, abnormal timing relationships may include, but are not limited to: timestamp reversal, missing timestamps, timestamps significantly deviating from a reasonable range, incorrect workstation sequence recording, and multiple records for the same pot body at the same workstation.

[0040] Correspondingly, data correction can include various correction methods such as timestamp correction, workstation identification correction, event node association correction, abnormal data annotation, and missing data supplementation.

[0041] In one feasible implementation, each production history trajectory can be traversed to check whether the timestamps of adjacent event nodes in the trajectory are in an increasing relationship. When a timestamp is found to be inverted, a reasonable timestamp value is calculated based on the flow time constraints of adjacent event nodes and the statistical regularity of other normal trajectories. For event nodes with incorrect workstation sequence records, their workstation identifiers are corrected according to the actual position of the node in the production history trajectory. For multiple records of the same pot body appearing at the same workstation, the validity of the record is determined based on time proximity and contextual consistency. Invalid records are marked as "duplicate records" or deleted, and the corrected data is updated in the production database.

[0042] By adopting the technical solution of this application, the dispersed production data of multiple workstations on a stainless steel pot production line within a target time period is converted into multiple event nodes, and the identity features of the stainless steel pots are extracted from them. In the case that a single pot lacks a continuously readable explicit identifier, implicit features are used to construct a distinguishable basis for each pot. Then, based on the process topology relationship, flow time constraints, and identity features between multiple workstations, candidate transfer relationships are established between event nodes to form a candidate transfer network. Under complex conditions such as buffer mixing, rework, order insertion disturbances, and data time asynchrony and delayed upload, multiple possible data attribution relationships are expressed in a structured way. By solving the trajectory of the candidate transfer network, the production history trajectory corresponding to each stainless steel pot is identified from multiple candidate paths, realizing the accurate attribution of dispersed production data to specific pot individuals, effectively restoring the complete production process of a single pot, and further improving the accuracy and consistency of the data by correcting production data with abnormal time sequence relationships.

[0043] Based on the above embodiments, as an optional embodiment, step S103 may further include the following steps.

[0044] Step S201: Based on the process topology, select event node pairs that meet the process sequence requirements to obtain the first candidate node pair; Step S202: Calculate the time difference between event nodes in the first candidate node pair; Step S203: The first candidate node pair whose time difference satisfies the corresponding flow time constraint is determined as the second candidate node pair; Step S204: Calculate the compatibility between the identity features of the event nodes in the second candidate node pair; Step S205: The second candidate node pair with a compatibility greater than the compatibility threshold is determined as a candidate transition edge; Step S206: Construct a candidate transition network by using multiple event nodes as nodes and candidate transition edges as edges.

[0045] In step S201, an event node pair refers to an ordered combination of two event nodes, where the first event node is the starting node and the second event node is the target node. A first candidate node pair refers to an event node pair where there is a process flow path between the workstation corresponding to the starting node and the workstation corresponding to the target node.

[0046] In one feasible implementation, a pre-stored process topology table can be obtained, which records the allowed flow relationships between each workstation. Multiple event nodes are traversed as starting nodes. For each starting node, event nodes whose workstation identifiers meet the process sequence requirements are selected from the remaining event nodes as target nodes. The starting node and the target node are combined to form an event node pair and recorded as the first candidate node pair.

[0047] Optionally, the process sequence requirements may include the sequential flow requirements of normal processes and the return path requirements of rework processes. Normal process requirements are met when the station of the target node is downstream of the station of the starting node, and rework process requirements are met when the station of the target node is in a preset set of rework target stations.

[0048] In step S202, the time difference refers to the difference between the timestamp of the target node and the timestamp of the starting node in the first candidate node pair.

[0049] In one feasible implementation, the timestamp fields of the starting node and the target node in the first candidate node pair can be read, the difference between the timestamp of the target node and the timestamp of the starting node can be calculated, and the difference can be used as the time difference of the first candidate node pair.

[0050] In step S203, the second candidate node pair refers to an event node pair that not only meets the process sequence requirements but also the requirements for reasonable turnaround time.

[0051] In one feasible implementation, the flow time constraint between the starting node workstation and the target node workstation can be obtained. The constraint includes the minimum flow time and the maximum flow time. It is determined whether the time difference of the first candidate node pair is greater than or equal to the minimum flow time and less than or equal to the maximum flow time. When the determination result is yes, the first candidate node pair is determined as the second candidate node pair. When the determination result is no, the first candidate node pair is eliminated.

[0052] It should be noted that due to the different physical distances and cache configurations between different workstations, the flow time constraints between different workstation pairs are different. When making a judgment, it is necessary to select the corresponding flow time constraint parameters according to the specific workstation pair.

[0053] In step S204, compatibility refers to the degree of consistency or matching between the identity features of the two event nodes in the second candidate node pair.

[0054] In one feasible implementation, the identity features of the starting node and the target node in the second candidate node pair can be extracted. For discrete features such as specifications, material type, and batch identification, it is determined whether the feature values ​​of the two nodes are completely identical. If they are identical, the compatibility of the feature item is recorded as 1; otherwise, it is recorded as 0. For continuous features such as pot body size parameters, the difference between the feature values ​​of the two nodes is calculated. When the difference is less than the preset allowable error, the compatibility of the feature item is recorded as 1; otherwise, it is recorded as 0. The compatibility of each feature item is weighted and summed to obtain the compatibility of the second candidate node pair.

[0055] Optionally, the compatibility calculation can also take into account the reliability of identity features, assigning higher weights to features from automatic identification devices and lower weights to features from manual input.

[0056] In step S205, each second candidate node pair is traversed, and a directed edge is established between the starting node and the target node in the second candidate node pair with a compatibility greater than the compatibility threshold. This directed edge is used as a candidate transition edge, and the second candidate node pairs with a compatibility less than or equal to the compatibility threshold are eliminated.

[0057] In step S206, for a candidate transfer network, an event node may have multiple incoming edges and multiple outgoing edges, reflecting the position of the event node in multiple possible pot body transfer paths. The subsequent trajectory solving process will select the optimal path combination from these candidate paths.

[0058] By adopting this implementation method, a candidate transfer network is constructed through a step-by-step screening process. First, a preliminary screening is performed based on the process topology relationship to obtain the first candidate node pair. Then, a time dimension screening is performed based on the flow time constraint to obtain the second candidate node pair. Finally, a refined screening is performed based on the identity feature compatibility to obtain the candidate transfer edges. While narrowing the candidate range layer by layer, reasonable candidate connections are retained, avoiding the omission of the real pot body flow relationship and avoiding the excessive computational burden caused by establishing too many unreasonable candidate connections. This improves the efficiency of trajectory solving while ensuring the accuracy of the candidate transfer network.

[0059] Based on the above embodiments, as an optional embodiment, the timing data alignment method for the stainless steel pot production line may further include the following steps.

[0060] Step S301: Obtain the standard production cycle time, the shortest dwell time and the longest waiting time of the buffer area between two adjacent workstations in the stainless steel pot production line; Step S302: Based on the standard production cycle time, the shortest dwell time, and the longest waiting time, calculate the minimum and maximum turnaround time between the two workstations. Step S303: The interval consisting of the minimum turnover time and the maximum turnover time is determined as the turnover time constraint between the two workstations.

[0061] In step S301, the standard production cycle time refers to the average time interval between two adjacent workstations during normal production of the stainless steel pot production line. The minimum dwell time refers to the minimum time the pot remains in the buffer area, which is usually determined by the physical transport time of the buffer area. The maximum waiting time refers to the maximum allowable dwell time of the pot in the buffer area; exceeding this time may lead to production abnormalities or material backlog.

[0062] In one feasible implementation, a pre-set standard production cycle value can be read from the process parameter configuration file of the production line. This cycle value is determined based on the equipment processing speed and the distance between workstations. The shortest dwell time and the longest waiting time can be obtained from the operating parameters of the buffer zone. The shortest dwell time includes the minimum physical time for the pot to enter the buffer zone, be transported within the buffer zone, and leave the buffer zone. The longest waiting time is set according to the buffer zone capacity and the buffering strategy of the production plan.

[0063] Optionally, the standard production cycle time can also be obtained by statistically analyzing the time difference between adjacent workstation event nodes in historical production data, and taking the median or mode of the time difference as the standard production cycle time.

[0064] In step S302, the minimum transfer time refers to the theoretical shortest time required for the pot body to transfer from the previous station to the next station, and the maximum transfer time refers to the longest time allowed for the pot body to transfer from the previous station to the next station.

[0065] It should be noted that when there is no buffer between two adjacent workstations, the minimum dwell time and the maximum waiting time are both set to zero. In this case, the minimum turnaround time and the maximum turnaround time are both equal to the standard production cycle time.

[0066] In step S303, the flow time constraint can be represented as a closed interval, with the lower bound of the interval being the minimum flow time and the upper bound of the interval being the maximum flow time.

[0067] In one feasible implementation, the numerical range consisting of the minimum turnaround time and the maximum turnaround time can be recorded as the turnaround time constraint of the workstation pair, and the constraint can be stored in the process topology relationship table and associated with the corresponding workstation pair. Subsequently, when screening candidate node pairs, the corresponding turnaround time constraint can be obtained by querying the table for time difference judgment.

[0068] By adopting this implementation method, the flow time constraint is calculated based on the actual operating parameters of the standard production cycle and the buffer zone. This ensures that the constraint parameters reflect both the standard production speed of the production line and the impact of the buffer zone on the flow time of the pot, thereby improving the rationality and accuracy of the flow time constraint.

[0069] Based on the above embodiments, as an optional embodiment, step S104 may further include the following steps.

[0070] Step S401: Calculate the continuity score of candidate transition edges in the candidate transition network; the continuity score characterizes the probability that two event nodes connected by the candidate transition edge belong to the same stainless steel pot body. Step S402: Based on continuity scoring and preset constraints, select target transfer edges from the candidate transfer network; the preset constraints are used to limit the uniqueness of the affiliation of each event node and the process rationality of the trajectory. Step S403: The path formed by the event nodes connected by the target transfer edge is determined as the production history trajectory of the corresponding stainless steel pot body.

[0071] In step S401, the continuity score refers to the credibility of the candidate transfer edge representing the flow relationship of the same pot body. The larger the value, the more likely the two event nodes connected by the candidate transfer edge belong to the continuous production process of the same pot body.

[0072] In one feasible implementation, for each candidate transfer edge, the time difference between the starting node and the target node connected by the edge, the compatibility of identity features, and the frequency of workstation transfer can be extracted. The closeness of the time difference to the standard production cycle can be calculated. The higher the closeness, the higher the time dimension score. The compatibility of identity features is used as the feature dimension score. The transfer frequency between the workstation pairs is counted from historical production data. The normalized transfer frequency is used as the process dimension score. The time dimension score, feature dimension score, and process dimension score are weighted and summed to obtain the continuity score of the candidate transfer edge.

[0073] In step S402, the preset constraints refer to the restrictions that need to be met during the trajectory solving process. The target transfer edge refers to the candidate transfer edge selected from the candidate transfer network that is determined to belong to the actual flow path of a certain pot body.

[0074] In one feasible implementation, preset constraints can be set, including node uniqueness constraints and path integrity constraints. The node uniqueness constraint requires that each event node can be used by at most one trajectory, and the path integrity constraint requires that each trajectory must contain a complete sequence of processes from the starting station to the ending station. The solution is obtained by using an integer programming algorithm or a maximum flow algorithm, with the objective function being to maximize the sum of continuity scores. Under the premise of satisfying the preset constraints, a set of candidate transition edges is selected as the target transition edge.

[0075] It should be noted that when there are multiple candidate transition edges pointing to the same event node in the candidate transition network, the node uniqueness constraint will ensure that only one candidate transition edge is selected and the rest of the candidate transition edges are eliminated, thereby avoiding the same event node being repeatedly assigned to different pots.

[0076] In step S403, in one feasible implementation, an event node sequence formed by continuous connection of the same starting node through the target transfer edge can be extracted from the target transfer edge set. Each event node sequence is regarded as a path, and each path corresponds to the production history trajectory of a stainless steel pot. The production history trajectory records the workstations that the pot passes through in sequence, the event node information of each workstation, and the transfer time between workstations.

[0077] Optionally, isolated event nodes that fail to connect into a complete path can be marked as unmatched nodes and recorded separately as objects for subsequent anomaly data analysis.

[0078] By adopting this implementation method, the credibility of different candidate paths is quantified by calculating the continuity score of candidate transfer edges. Combined with preset constraints, the optimal combination of target transfer edges is selected from the candidate transfer network. Under the premise of ensuring the unique attribution of each event node and the rationality of the trajectory process, the actual production history trajectory of each pot body can be accurately identified from multiple candidate paths.

[0079] Based on the above embodiments, as an optional embodiment, step S401 may further include the following steps.

[0080] Obtain the temporal matching degree, identity feature compatibility, process stage continuity, and spatial location migration degree between the two event nodes connected by the candidate transition edge; The continuity score of candidate transition edges is obtained by weighted summation of time matching degree, identity feature compatibility, process stage continuity and spatial location mobility.

[0081] Among them, the time matching degree characterizes the degree of conformity between the time interval of two event nodes and the corresponding workstation flow time constraint; the process stage continuity characterizes the rationality of the process stage advancement corresponding to the two event nodes; and the spatial location mobility characterizes the physical accessibility of the changes in the cache location or vehicle location involved in the two event nodes.

[0082] In one feasible implementation, the calculation of time matching degree specifically includes: obtaining the time difference between the starting node and the target node of the candidate transfer edge connection, and the flow time constraint between the starting node workstation and the target node workstation, calculating the degree of deviation between the time difference and the median of the flow time constraint, and the smaller the degree of deviation, the higher the time matching degree.

[0083] Optionally, when the time difference is within the interval of the flow time constraint, a basic time matching degree can be assigned, and then refined according to the position of the time difference within the interval. The matching degree is highest when it is in the middle of the interval, and relatively low when it is at the boundary of the interval.

[0084] The calculation of process stage continuity specifically includes: obtaining the process stage identifiers corresponding to the starting node and target node connected by the candidate transition edge, determining the advancement relationship between the two process stages, setting the process stage continuity to the first value when the process stage of the target node is a direct successor to the process stage of the starting node, setting the process stage continuity to the second value when the process stage of the target node is an indirect successor to the process stage of the starting node, and setting the process stage continuity to the third value when the process stage of the target node is a reflow stage of the process stage of the starting node. The first value is greater than the second value, and the second value is greater than the third value.

[0085] For example, the production of stainless steel pots can be divided into the billet preparation stage, the forming and processing stage, the surface treatment stage, and the quality inspection and packaging stage. When the starting node belongs to the forming and processing stage and the target node belongs to the surface treatment stage, the two are directly related and the process stage continuity is set to 1. When the starting node belongs to the surface treatment stage and the target node belongs to the forming and processing stage, they are related to rework and reflow and the process stage continuity is set to 0.3.

[0086] The calculation of spatial location mobility specifically includes: obtaining the spatial location information corresponding to the starting node and target node of the candidate transfer edge connection. The spatial location information includes the buffer number, buffer bit number or vehicle number. The physical connectivity and transmission distance between the two spatial locations are determined. When there is a direct transmission channel between the two spatial locations, the spatial location mobility is set to a higher value. When the two spatial locations need to go through a transfer, the spatial location mobility decreases according to the number of transfers. When there is no physical connectivity between the two spatial locations, the spatial location mobility is set to zero.

[0087] Optionally, when an event node does not record spatial location information, the physical distance between workstations can be used as a substitute; the closer the distance, the higher the degree of spatial location migration.

[0088] By adopting this implementation method, the time matching degree measures the rationality of the transfer time, the identity feature compatibility degree measures the consistency of the pot body attributes, the process stage continuity degree measures the sequentiality of the process advancement, and the spatial location migration degree measures the feasibility of physical transfer. By weighted summation, the multi-dimensional information is integrated into a unified continuity score, so that the score comprehensively and accurately reflects the credibility of the candidate transfer edge.

[0089] Based on the above embodiments, as an optional embodiment, the preset constraints include at least one or more of the following combinations: The first constraint is that each event node belongs to at most one production history trajectory. This first constraint is used to ensure that the same event node will not be repeatedly assigned to multiple different pots, thus avoiding logical contradictions caused by overlapping trajectories.

[0090] The second constraint is that each event node has at most one predecessor event node and one successor event node. This second constraint is used to ensure that the production history trajectory forms a single continuous path, and each event node has only a unique upstream node and downstream node in the trajectory, so as to avoid the trajectory from branching or merging.

[0091] The third constraint is that the sequence of event nodes in the production history trajectory conforms to the process order in the process topology. The third constraint is used to ensure that the order of workstations in the production history trajectory conforms to the predefined process topology. For example, the surface polishing process must be after the forming and stamping process and before the quality inspection process. No reverse order or jumps that violate the process logic are allowed.

[0092] The fourth constraint is that the number of stainless steel pots in the buffer zone at any given time does not exceed the upper limit of the buffer zone capacity. This fourth constraint is used to limit the number of work-in-process items in the buffer zone at the same time. During the solution process, if a target transfer edge is selected that causes the number of pots in the buffer zone at a certain time to exceed the upper limit of the capacity, the target transfer edge cannot be selected. This constraint is used to prevent the solution result from exceeding the actual carrying capacity of the physical space.

[0093] The fifth constraint is that the difference between the number of production history trajectories and the output statistics within the target period must be within a preset range. This fifth constraint is used to verify the total number of trajectories obtained by the solution. It compares the number of production history trajectories obtained by the solution with the actual output quantity obtained from the production management system within the target period. When the difference exceeds the preset range, it indicates that the solution results may contain omissions or redundancies, and the calculation parameters or constraints of the continuity score need to be adjusted and the solution re-solved.

[0094] It should be noted that in practical applications, the above constraints can be flexibly combined according to the solution effect and the actual production situation. For scenarios with small production line scale and good data quality, only the first three constraints can be used. For scenarios with high production line complexity and multiple buffers, it is recommended to use all five constraints to improve the accuracy of the solution.

[0095] By adopting this implementation method, the preset constraints restrict the trajectory solution from multiple perspectives, such as node uniqueness, path continuity, process rationality, physical feasibility, and quantity consistency, and eliminate candidate solutions that do not conform to production reality during the target transfer edge selection process.

[0096] Based on the above embodiments, as an optional embodiment, step S105 may further include the following steps.

[0097] Step S501: Determine the sequential relationship of event nodes in the production history trajectory as the standard time sequence relationship of the corresponding stainless steel pot body. Step S502: Detect data items in the original production data that are inconsistent with the standard time series relationship and identify them as abnormal time series data; abnormal time series data includes abnormal timestamp data and abnormal identity data; Step S503: Based on the information of the corresponding event nodes in the production history trajectory, correct the abnormal time series data to obtain the corrected production data. Step S504: Sort the corrected production data according to the standard time sequence relationship.

[0098] In step S501, the standard timing relationship refers to the sequence of events and time constraints that are determined through trajectory solving and conform to actual production.

[0099] In one feasible implementation, for each production history trajectory, the workstation sequence, timestamp sequence, and identity feature sequence of the event nodes in the trajectory can be extracted. The event nodes are established in chronological order according to their positions in the trajectory, and the time difference between adjacent event nodes is used as the flow time reference value. This chronological order and time reference value are recorded as the standard time sequence relationship of the pot body.

[0100] In step S502, each data item in the original production data can be traversed, the corresponding production history trajectory can be located according to the identity identifier in the data item, the proper position of the data item in the standard time sequence relationship can be queried, whether the timestamp of the data item is consistent with the time constraint in the standard time sequence relationship can be checked, and whether the identity identifier of the data item is consistent with the identity characteristics of the corresponding event node in the production history trajectory can be checked. When the timestamp or identity identifier is inconsistent, the data item is marked as abnormal time sequence data.

[0101] For example, the standard timing relationship of a pot body at the polishing station shows that the timestamp of the preceding stamping station is 10:00 and the timestamp of the subsequent inspection station is 10:30. However, the timestamp of the polishing station in the original data is recorded as 09:50, which is earlier than the stamping station time. Therefore, the timestamp of the polishing station data item is abnormal data.

[0102] For detecting abnormal identity data, the identity characteristics of the event node corresponding to the data item in the production history can be obtained. The identity identifier recorded in the data item is compared with the identity characteristics. When the key fields of the identity identifier do not match the identity characteristics, it is determined that the identity identifier is abnormal.

[0103] For example, if the production history shows that a certain pot is a 24cm non-stick pot, while the original data shows that the identification of a certain workstation is a 28cm stainless steel pot, the two do not match, then the identification of this data item is abnormal data.

[0104] In step S503, for abnormal timestamp data, the timestamp of the event node corresponding to the data item in the production history trajectory can be obtained, and the abnormal timestamp in the original data item can be replaced with the timestamp. For abnormal identity data, the identity feature of the event node corresponding to the data item in the production history trajectory can be obtained, and the identity feature can be used to reconstruct the identity field of the data item.

[0105] Optionally, when correcting abnormal time-series data, a correction flag field can be added to the corrected production data to record the correction process and the specific details of the correction.

[0106] In step S504, all data items corresponding to the same production history trajectory can be arranged in the order of the standard time sequence relationship, with the timestamp as the primary key and the process sequence as the secondary key, to obtain ordered production data organized according to the standard time sequence relationship.

[0107] In one feasible implementation, a unique trajectory identifier can be assigned to each production history trajectory, the trajectory identifier can be added to the corresponding data item, and a two-level sorting can be performed according to the trajectory identifier and the order within the trajectory, so that all production data of the same pot are arranged continuously, and the production data of different pots are grouped according to the trajectory identifier.

[0108] By adopting this implementation method, the production history trajectory is converted into a standard time sequence relationship as a reference benchmark for data correction. By comparing the original data with the standard time sequence relationship, anomalies in timestamps and identification are identified. The accurate information in the production history trajectory is used to correct the abnormal data. Finally, the data is reordered according to the standard time sequence relationship, thus completing the transformation from the original messy data to the ordered and accurate data.

[0109] Based on the above embodiments, as an optional embodiment, the identity features of the stainless steel pot body include one or more of the following combinations: The specifications and models of stainless steel pot bodies indicate the size type and product series of the pot body. The material type of the stainless steel pot body, which indicates the grade and thickness of the steel used in the pot body; The production batch mark of the stainless steel pot body indicates the production batch and the time period of material feeding of the pot body. The surface treatment type of stainless steel pot body, which describes the treatment method of the pot body surface; The accessory configuration information for the stainless steel pot body includes the type of handle, lid, and lugs provided with the pot body.

[0110] It should be noted that the selection of identity features can be adjusted according to the actual records in the production data and business needs. For production lines with a high degree of informatization, all five identity features can be used to achieve refined pot body identification. For production lines with simpler data records, only the two core features of specifications and production batch identification can be used.

[0111] By adopting this implementation method, the identity characteristics of the pot body are defined from multiple dimensions such as specifications, materials, batches, processes and configurations. When calculating the compatibility of identity characteristics, the matching degree of multiple dimensions can be comprehensively considered, thereby improving the accuracy of pot body identity recognition.

[0112] Figure 2 A schematic diagram of the structure of a timing data alignment system for a stainless steel pot production line provided in this application embodiment is shown below. Figure 2 As shown, the timing data alignment system of this stainless steel pot production line includes: The event node generation module is used to acquire production data from multiple workstations on the stainless steel pot production line within a target time period and convert the production data into multiple event nodes. The identity feature extraction module is used to extract the identity features of the stainless steel pot body from multiple event nodes; The candidate network construction module is used to establish candidate transfer relationships between multiple event nodes based on the process topology relationship, flow time constraints and identity characteristics between multiple workstations, forming a candidate transfer network; The production trajectory solving module is used to solve the trajectory of the candidate transfer network and obtain the production history trajectory of each of the multiple stainless steel pots within the target time period. The time-series data alignment module is used to correct production data with abnormal time-series relationships within a target time period based on the production history trajectory.

[0113] Based on the above embodiments, as an optional embodiment, the candidate network construction module is further configured to: filter out event node pairs that meet the process sequence requirements according to the process topology relationship to obtain a first candidate node pair; calculate the time difference between event nodes in the first candidate node pair; determine the first candidate node pair whose time difference meets the corresponding flow time constraint as a second candidate node pair; calculate the compatibility between the identity features of event nodes in the second candidate node pair; determine the second candidate node pair with a compatibility greater than the compatibility threshold as a candidate transition edge; and construct a candidate transition network by using multiple event nodes as nodes and candidate transition edges as edges.

[0114] Based on the above embodiments, as an optional embodiment, the candidate network construction module is also used to obtain the standard production cycle time, the shortest dwell time of the buffer area, and the longest waiting time between two adjacent workstations in the stainless steel pot production line; calculate the minimum turnover time and the maximum turnover time between the two workstations based on the standard production cycle time, the shortest dwell time, and the longest waiting time; and determine the interval composed of the minimum turnover time and the maximum turnover time as the turnover time constraint between the two workstations.

[0115] Based on the above embodiments, as an optional embodiment, the production trajectory solving module is also used to calculate the continuity score of candidate transfer edges in the candidate transfer network; the continuity score characterizes the probability that two event nodes connected by the candidate transfer edge belong to the same stainless steel pot body; based on the continuity score and preset constraints, a target transfer edge is selected from the candidate transfer network; the preset constraints are used to limit the uniqueness of the affiliation of each event node and the process rationality of the trajectory; the path formed by the event nodes connected by the target transfer edge is determined as the production history trajectory of the corresponding stainless steel pot body.

[0116] Based on the above embodiments, as an optional embodiment, the production trajectory solving module is also used to obtain the time matching degree, identity feature compatibility, process stage continuity and spatial location migration degree between the two event nodes connected by the candidate transfer edge; and to obtain the continuity score of the candidate transfer edge by weighted summation of the time matching degree, identity feature compatibility, process stage continuity and spatial location migration degree.

[0117] Based on the above embodiments, as an optional embodiment, the time-series data alignment module is further used to determine the sequential relationship of event nodes in the production history trajectory as the standard time-series relationship corresponding to the stainless steel pot body; detect data items in the original production data that are inconsistent with the standard time-series relationship and determine them as abnormal time-series data; abnormal time-series data includes abnormal timestamp data and abnormal identity data; correct the abnormal time-series data based on the information of the corresponding event nodes in the production history trajectory to obtain corrected production data; and sort the corrected production data according to the standard time-series relationship.

[0118] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of this application, such as... Figure 3 As shown, the electronic device may include a processor 310, a communications interface 320, a memory 330, and a communication bus 340. The processor 310, communications interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions from the memory 330 to execute the timing data alignment method for the stainless steel pot production line.

[0119] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0120] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the timing data alignment method for the stainless steel pot production line provided by the above methods.

[0121] In another aspect, this application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the timing data alignment method for the stainless steel pot production line provided by the above methods.

[0122] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0123] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A timing data alignment method for a stainless steel pot production line, characterized by, include: Acquire production data from multiple workstations on the stainless steel pot production line within a target time period, and convert the production data into multiple event nodes; Extract the identity features of the stainless steel pot body from the multiple event nodes; Based on the process topology relationship, flow time constraints and identity characteristics among the multiple workstations, candidate transfer relationships are established among the multiple event nodes to form a candidate transfer network. The candidate transfer network is used to solve for the trajectory, and the production history trajectory of each of the multiple stainless steel pots within the target time period is obtained. Based on the production history trajectory, production data with abnormal time sequence relationships within the target time period are corrected.

2. The timing data alignment method of a stainless steel pot production line according to claim 1, characterized by, Based on the process topology relationships, flow time constraints, and identity characteristics among the multiple workstations, candidate transfer relationships are established among the multiple event nodes to form a candidate transfer network, including: Based on the process topology, event node pairs that meet the process sequence requirements are selected to obtain the first candidate node pairs; Calculate the time difference between event nodes in the first candidate node pair; The first candidate node pair whose time difference satisfies the corresponding flow time constraint is determined as the second candidate node pair; Calculate the compatibility between the identity features of the event nodes in the second candidate node pair; The second candidate node pair with a compatibility greater than the compatibility threshold is determined as a candidate transition edge; The candidate transition network is constructed by using the multiple event nodes as nodes and the candidate transition edges as edges.

3. The timing data alignment method of a stainless steel pot production line according to claim 2, characterized by, The method further includes: Obtain the standard production cycle time, the shortest dwell time in the buffer zone, and the longest waiting time between two adjacent workstations in the stainless steel pot production line; Based on the standard production cycle time, the shortest dwell time, and the longest waiting time, calculate the minimum and maximum turnaround times between the two workstations. The interval consisting of the minimum and maximum turnover times is defined as the turnover time constraint between the two workstations.

4. The timing data alignment method of a stainless steel pot production line according to claim 1, characterized by, The process of solving the trajectory of the candidate transfer network to obtain the production history trajectory of each of the multiple stainless steel pots within the target time period includes: Calculate the continuity score of the candidate transition edges in the candidate transition network; the continuity score characterizes the degree of probability that the two event nodes connected by the candidate transition edge belong to the same stainless steel pot body; Based on the continuity score and preset constraints, a target transfer edge is selected from the candidate transfer network; the preset constraints are used to limit the uniqueness of the affiliation of each event node and the process rationality of the trajectory. The path formed by connecting the event nodes of the target transfer edge is determined as the production history trajectory of the corresponding stainless steel pot body.

5. The timing data alignment method of a stainless steel pot production line according to claim 4, characterized by, The calculation of the continuity score of candidate transition edges in the candidate transition network includes: Obtain the temporal matching degree, identity feature compatibility, process stage continuity, and spatial location migration degree between the two event nodes connected by the candidate transition edge; The continuity score of the candidate transition edge is obtained by weighted summation of the time matching degree, identity feature compatibility, process stage continuity and spatial location mobility. The time matching degree represents the degree to which the time interval between two event nodes conforms to the corresponding workstation flow time constraint; The continuity of the process stages characterizes the rationality of the process stage advancement corresponding to two event nodes; The spatial location mobility characterizes the physical reachability of changes in cache location or vehicle location involved between two event nodes.

6. The timing data alignment method of a stainless steel pot production line according to claim 4, characterized by, The preset constraints include at least one or more of the following combinations: Each event node can belong to at most one production history track; Each event node has at most one predecessor event node and one successor event node; The sequence of event nodes in the production history trajectory conforms to the process order in the process topology; At any given time, the number of stainless steel pot bodies under manufacturing in the buffer area shall not exceed the upper limit of the buffer area capacity. The difference between the number of production history trajectories and the output statistics within the target period is within the preset range.

7. The time-series data alignment method for a stainless steel pot production line according to claim 1, characterized in that, The step of correcting production data with abnormal temporal relationships within the target time period based on the production history trajectory includes: The sequential relationship of event nodes in the production history trajectory is determined as the standard temporal relationship of the corresponding stainless steel pot body; Data items in the original production data that are inconsistent with the standard time series relationship are identified as abnormal time series data; the abnormal time series data includes abnormal timestamp data and abnormal identity identification data. Based on the information of the corresponding event nodes in the production history trajectory, the abnormal time series data is corrected to obtain the corrected production data. The corrected production data is sorted according to the standard time sequence relationship.

8. The time-series data alignment method for a stainless steel pot production line according to claim 1, characterized in that, The identification characteristics of the stainless steel pot body include one or more of the following: The specifications and models of the stainless steel pot body, wherein the specifications and models represent the size type and product series of the pot body; The material type of the stainless steel pot body, wherein the material type characterizes the steel grade and thickness level of the raw materials used in the pot body; The production batch identifier of the stainless steel pot body, wherein the production batch identifier indicates the production batch to which the pot body belongs and the time period of material feeding; The surface treatment type of the stainless steel pot body, wherein the surface treatment type characterizes the treatment method of the pot body surface; The accessory configuration information for the stainless steel pot body includes the type of handle, lid, and lugs provided with the pot body.

9. A timing data alignment system for a stainless steel pot production line, characterized in that, include: The event node generation module is used to acquire production data from multiple workstations on the stainless steel pot production line within a target time period and convert the production data into multiple event nodes. An identity feature extraction module is used to extract the identity features of the stainless steel pot body from the multiple event nodes; The candidate network construction module is used to establish candidate transfer relationships among the multiple event nodes based on the process topology relationship, flow time constraints and identity characteristics among the multiple workstations, to form a candidate transfer network. The production trajectory solving module is used to solve the trajectory of the candidate transfer network to obtain the production history trajectory of each of the multiple stainless steel pots within the target time period. The time-series data alignment module is used to correct production data with abnormal time-series relationships within the target time period based on the production history trajectory.

10. 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 computer program, it implements the timing data alignment method for the stainless steel pot production line as described in any one of claims 1-8.

Images